20220308.md

ArXiv cs.CV --Tue, 8 Mar 2022

1.ZippyPoint: Fast Interest Point Detection, Description, and Matching through Mixed Precision Discretization ⬇️

The design of more complex and powerful neural network models has significantly advanced the state-of-the-art in local feature detection and description. These advances can be attributed to deeper networks, improved training methodologies through self-supervision, or the introduction of new building blocks, such as graph neural networks for feature matching. However, in the pursuit of increased performance, efficient architectures that generate lightweight descriptors have received surprisingly little attention. In this paper, we investigate the adaptations neural networks for detection and description require in order to enable their use in embedded platforms. To that end, we investigate and adapt network quantization techniques for use in real-time applications. In addition, we revisit common practices in descriptor quantization and propose the use of a binary descriptor normalization layer, enabling the generation of distinctive length-invariant binary descriptors. ZippyPoint, our efficient network, runs at 47.2 fps on the Apple M1 CPU. This is up to 5x faster than other learned detection and description models, making it the only real-time learned network. ZippyPoint consistently outperforms all other binary detection and descriptor methods in visual localization and homography estimation tasks. Code and trained models will be released upon publication.

2.Human-Aware Object Placement for Visual Environment Reconstruction ⬇️

Humans are in constant contact with the world as they move through it and interact with it. This contact is a vital source of information for understanding 3D humans, 3D scenes, and the interactions between them. In fact, we demonstrate that these human-scene interactions (HSIs) can be leveraged to improve the 3D reconstruction of a scene from a monocular RGB video. Our key idea is that, as a person moves through a scene and interacts with it, we accumulate HSIs across multiple input images, and optimize the 3D scene to reconstruct a consistent, physically plausible and functional 3D scene layout. Our optimization-based approach exploits three types of HSI constraints: (1) humans that move in a scene are occluded or occlude objects, thus, defining the depth ordering of the objects, (2) humans move through free space and do not interpenetrate objects, (3) when humans and objects are in contact, the contact surfaces occupy the same place in space. Using these constraints in an optimization formulation across all observations, we significantly improve the 3D scene layout reconstruction. Furthermore, we show that our scene reconstruction can be used to refine the initial 3D human pose and shape (HPS) estimation. We evaluate the 3D scene layout reconstruction and HPS estimation qualitatively and quantitatively using the PROX and PiGraphs datasets. The code and data are available for research purposes at this https URL.

3.DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection ⬇️

We present DINO (\textbf{D}ETR with \textbf{I}mproved de\textbf{N}oising anch\textbf{O}r boxes), a state-of-the-art end-to-end object detector. % in this paper. DINO improves over previous DETR-like models in performance and efficiency by using a contrastive way for denoising training, a mixed query selection method for anchor initialization, and a look forward twice scheme for box prediction. DINO achieves $48.3$AP in $12$ epochs and $51.0$AP in $36$ epochs on COCO with a ResNet-50 backbone and multi-scale features, yielding a significant improvement of $\textbf{+4.9}$\textbf{AP} and $\textbf{+2.4}$\textbf{AP}, respectively, compared to DN-DETR, the previous best DETR-like model. DINO scales well in both model size and data size. Without bells and whistles, after pre-training on the Objects365 dataset with a SwinL backbone, DINO obtains the best results on both COCO \texttt{val2017} ($\textbf{63.2}$\textbf{AP}) and \texttt{test-dev} (\textbf{$\textbf{63.3}$AP}). Compared to other models on the leaderboard, DINO significantly reduces its model size and pre-training data size while achieving better results. Our code will be available at \url{this https URL}.

4.Audio-visual Generalised Zero-shot Learning with Cross-modal Attention and Language ⬇️

Learning to classify video data from classes not included in the training data, i.e. video-based zero-shot learning, is challenging. We conjecture that the natural alignment between the audio and visual modalities in video data provides a rich training signal for learning discriminative multi-modal representations. Focusing on the relatively underexplored task of audio-visual zero-shot learning, we propose to learn multi-modal representations from audio-visual data using cross-modal attention and exploit textual label embeddings for transferring knowledge from seen classes to unseen classes. Taking this one step further, in our generalised audio-visual zero-shot learning setting, we include all the training classes in the test-time search space which act as distractors and increase the difficulty while making the setting more realistic. Due to the lack of a unified benchmark in this domain, we introduce a (generalised) zero-shot learning benchmark on three audio-visual datasets of varying sizes and difficulty, VGGSound, UCF, and ActivityNet, ensuring that the unseen test classes do not appear in the dataset used for supervised training of the backbone deep models. Comparing multiple relevant and recent methods, we demonstrate that our proposed AVCA model achieves state-of-the-art performance on all three datasets. Code and data will be available at \url{this https URL}.

5.On the pitfalls of entropy-based uncertainty for multi-class semi-supervised segmentation ⬇️

Semi-supervised learning has emerged as an appealing strategy to train deep models with limited supervision. Most prior literature under this learning paradigm resorts to dual-based architectures, typically composed of a teacher-student duple. To drive the learning of the student, many of these models leverage the aleatoric uncertainty derived from the entropy of the predictions. While this has shown to work well in a binary scenario, we demonstrate in this work that this strategy leads to suboptimal results in a multi-class context, a more realistic and challenging setting. We argue, indeed, that these approaches underperform due to the erroneous uncertainty approximations in the presence of inter-class overlap. Furthermore, we propose an alternative solution to compute the uncertainty in a multi-class setting, based on divergence distances and which account for inter-class overlap. We evaluate the proposed solution on a challenging multi-class segmentation dataset and in two well-known uncertainty-based segmentation methods. The reported results demonstrate that by simply replacing the mechanism used to compute the uncertainty, our proposed solution brings substantial improvement on tested setups.

6.The Unsurprising Effectiveness of Pre-Trained Vision Models for Control ⬇️

Recent years have seen the emergence of pre-trained representations as a powerful abstraction for AI applications in computer vision, natural language, and speech. However, policy learning for control is still dominated by a tabula-rasa learning paradigm, with visuo-motor policies often trained from scratch using data from deployment environments. In this context, we revisit and study the role of pre-trained visual representations for control, and in particular representations trained on large-scale computer vision datasets. Through extensive empirical evaluation in diverse control domains (Habitat, DeepMind Control, Adroit, Franka Kitchen), we isolate and study the importance of different representation training methods, data augmentations, and feature hierarchies. Overall, we find that pre-trained visual representations can be competitive or even better than ground-truth state representations to train control policies. This is in spite of using only out-of-domain data from standard vision datasets, without any in-domain data from the deployment environments. Additional details and source code is available at this https URL

7.Kubric: A scalable dataset generator ⬇️

Data is the driving force of machine learning, with the amount and quality of training data often being more important for the performance of a system than architecture and training details. But collecting, processing and annotating real data at scale is difficult, expensive, and frequently raises additional privacy, fairness and legal concerns. Synthetic data is a powerful tool with the potential to address these shortcomings: 1) it is cheap 2) supports rich ground-truth annotations 3) offers full control over data and 4) can circumvent or mitigate problems regarding bias, privacy and licensing. Unfortunately, software tools for effective data generation are less mature than those for architecture design and training, which leads to fragmented generation efforts. To address these problems we introduce Kubric, an open-source Python framework that interfaces with PyBullet and Blender to generate photo-realistic scenes, with rich annotations, and seamlessly scales to large jobs distributed over thousands of machines, and generating TBs of data. We demonstrate the effectiveness of Kubric by presenting a series of 13 different generated datasets for tasks ranging from studying 3D NeRF models to optical flow estimation. We release Kubric, the used assets, all of the generation code, as well as the rendered datasets for reuse and modification.

8.An Unsupervised Domain Adaptive Approach for Multimodal 2D Object Detection in Adverse Weather Conditions ⬇️

Integrating different representations from complementary sensing modalities is crucial for robust scene interpretation in autonomous driving. While deep learning architectures that fuse vision and range data for 2D object detection have thrived in recent years, the corresponding modalities can degrade in adverse weather or lighting conditions, ultimately leading to a drop in performance. Although domain adaptation methods attempt to bridge the domain gap between source and target domains, they do not readily extend to heterogeneous data distributions. In this work, we propose an unsupervised domain adaptation framework, which adapts a 2D object detector for RGB and lidar sensors to one or more target domains featuring adverse weather conditions. Our proposed approach consists of three components. First, a data augmentation scheme that simulates weather distortions is devised to add domain confusion and prevent overfitting on the source data. Second, to promote cross-domain foreground object alignment, we leverage the complementary features of multiple modalities through a multi-scale entropy-weighted domain discriminator. Finally, we use carefully designed pretext tasks to learn a more robust representation of the target domain data. Experiments performed on the DENSE dataset show that our method can substantially alleviate the domain gap under the single-target domain adaptation (STDA) setting and the less explored yet more general multi-target domain adaptation (MTDA) setting.

9.Weakly Supervised Learning of Keypoints for 6D Object Pose Estimation ⬇️

State-of-the-art approaches for 6D object pose estimation require large amounts of labeled data to train the deep networks. However, the acquisition of 6D object pose annotations is tedious and labor-intensive in large quantity. To alleviate this problem, we propose a weakly supervised 6D object pose estimation approach based on 2D keypoint detection. Our method trains only on image pairs with known relative transformations between their viewpoints. Specifically, we assign a set of arbitrarily chosen 3D keypoints to represent each unknown target 3D object and learn a network to detect their 2D projections that comply with the relative camera viewpoints. During inference, our network first infers the 2D keypoints from the query image and a given labeled reference image. We then use these 2D keypoints and the arbitrarily chosen 3D keypoints retained from training to infer the 6D object pose. Extensive experiments demonstrate that our approach achieves comparable performance with state-of-the-art fully supervised approaches.

10.Towards Unbiased Multi-label Zero-Shot Learning with Pyramid and Semantic Attention ⬇️

Multi-label zero-shot learning extends conventional single-label zero-shot learning to a more realistic scenario that aims at recognizing multiple unseen labels of classes for each input sample. Existing works usually exploit attention mechanism to generate the correlation among different labels. However, most of them are usually biased on several major classes while neglect most of the minor classes with the same importance in input samples, and may thus result in overly diffused attention maps that cannot sufficiently cover minor classes. We argue that disregarding the connection between major and minor classes, i.e., correspond to the global and local information, respectively, is the cause of the problem. In this paper, we propose a novel framework of unbiased multi-label zero-shot learning, by considering various class-specific regions to calibrate the training process of the classifier. Specifically, Pyramid Feature Attention (PFA) is proposed to build the correlation between global and local information of samples to balance the presence of each class. Meanwhile, for the generated semantic representations of input samples, we propose Semantic Attention (SA) to strengthen the element-wise correlation among these vectors, which can encourage the coordinated representation of them. Extensive experiments on the large-scale multi-label zero-shot benchmarks NUS-WIDE and Open-Image demonstrate that the proposed method surpasses other representative methods by significant margins.

11.Multi-Modal Attribute Extraction for E-Commerce ⬇️

To improve users' experience as they navigate the myriad of options offered by online marketplaces, it is essential to have well-organized product catalogs. One key ingredient to that is the availability of product attributes such as color or material. However, on some marketplaces such as Rakuten-Ichiba, which we focus on, attribute information is often incomplete or even missing. One promising solution to this problem is to rely on deep models pre-trained on large corpora to predict attributes from unstructured data, such as product descriptive texts and images (referred to as modalities in this paper). However, we find that achieving satisfactory performance with this approach is not straightforward but rather the result of several refinements, which we discuss in this paper. We provide a detailed description of our approach to attribute extraction, from investigating strong single-modality methods, to building a solid multimodal model combining textual and visual information. One key component of our multimodal architecture is a novel approach to seamlessly combine modalities, which is inspired by our single-modality investigations. In practice, we notice that this new modality-merging method may suffer from a modality collapse issue, i.e., it neglects one modality. Hence, we further propose a mitigation to this problem based on a principled regularization scheme. Experiments on Rakuten-Ichiba data provide empirical evidence for the benefits of our approach, which has been also successfully deployed to Rakuten-Ichiba. We also report results on publicly available datasets showing that our model is competitive compared to several recent multimodal and unimodal baselines.

12.Depth-SIMS: Semi-Parametric Image and Depth Synthesis ⬇️

In this paper we present a compositing image synthesis method that generates RGB canvases with well aligned segmentation maps and sparse depth maps, coupled with an in-painting network that transforms the RGB canvases into high quality RGB images and the sparse depth maps into pixel-wise dense depth maps. We benchmark our method in terms of structural alignment and image quality, showing an increase in mIoU over SOTA by 3.7 percentage points and a highly competitive FID. Furthermore, we analyse the quality of the generated data as training data for semantic segmentation and depth completion, and show that our approach is more suited for this purpose than other methods.

13.Screentone-Preserved Manga Retargeting ⬇️

As a popular comic style, manga offers a unique impression by utilizing a rich set of bitonal patterns, or screentones, for illustration. However, screentones can easily be contaminated with visual-unpleasant aliasing and/or blurriness after resampling, which harms its visualization on displays of diverse resolutions. To address this problem, we propose the first manga retargeting method that synthesizes a rescaled manga image while retaining the screentone in each screened region. This is a non-trivial task as accurate region-wise segmentation remains challenging. Fortunately, the rescaled manga shares the same region-wise screentone correspondences with the original manga, which enables us to simplify the screentone synthesis problem as an anchor-based proposals selection and rearrangement problem. Specifically, we design a novel manga sampling strategy to generate aliasing-free screentone proposals, based on hierarchical grid-based anchors that connect the correspondences between the original and the target rescaled manga. Furthermore, a Recurrent Proposal Selection Module (RPSM) is proposed to adaptively integrate these proposals for target screentone synthesis. Besides, to deal with the translation insensitivity nature of screentones, we propose a translation-invariant screentone loss to facilitate the training convergence. Extensive qualitative and quantitative experiments are conducted to verify the effectiveness of our method, and notably compelling results are achieved compared to existing alternative techniques.

14.A Glyph-driven Topology Enhancement Network for Scene Text Recognition ⬇️

Attention-based methods by establishing one-dimensional (1D) and two-dimensional (2D) mechanisms with an encoder-decoder framework have dominated scene text recognition (STR) tasks due to their capabilities of building implicit language representations. However, 1D attention-based mechanisms suffer from alignment drift on latter characters. 2D attention-based mechanisms only roughly focus on the spatial regions of characters without excavating detailed topological structures, which reduces the visual performance. To mitigate the above issues, we propose a novel Glyph-driven Topology Enhancement Network (GTEN) to improve topological features representations in visual models for STR. Specifically, an unsupervised method is first employed to exploit 1D sequence-aligned attention weights. Second, we construct a supervised segmentation module to capture 2D ordered and pixel-wise topological information of glyphs without extra character-level annotations. Third, these resulting outputs fuse enhanced topological features to enrich semantic feature representations for STR. Experiments demonstrate that GTEN achieves competitive performance on IIIT5K-Words, Street View Text, ICDAR-series, SVT Perspective, and CUTE80 datasets.

15.Spatio-temporal Gait Feature with Adaptive Distance Alignment ⬇️

Gait recognition is an important recognition technology, because it is not easy to camouflage and does not need cooperation to recognize subjects. However, there are still serious challenges in gait recognition, that is, people with similar walking posture are often recognized incorrectly. In this paper, We try to increase the difference of gait features of different subjects from two aspects: the optimization of network structure and the refinement of extracted gait features, so as to increase the recognition efficiency of subjects with similar walking posture. So our method is proposed, it consists of Spatio-temporal Feature Extraction (SFE) and Adaptive Distance Alignment (ADA), which SFE uses Temporal Feature Fusion (TFF) and Fine-grained Feature Extraction (FFE) to effectively extract the spatio-temporal features from raw silhouettes, ADA uses a large number of unlabeled gait data in real life as a benchmark to refine the extracted spatio-temporal features to make them have low inter-class similarity and high intra-class similarity. Extensive experiments on mini-OUMVLP and CASIA-B have proved that we have a good result than some state-of-the-art methods.

16.Adversarial Texture for Fooling Person Detectors in the Physical World ⬇️

Nowadays, cameras equipped with AI systems can capture and analyze images to detect people automatically. However, the AI system can make mistakes when receiving deliberately designed patterns in the real world, i.e., physical adversarial examples. Prior works have shown that it is possible to print adversarial patches on clothes to evade DNN-based person detectors. However, these adversarial examples could have catastrophic drops in the attack success rate when the viewing angle (i.e., the camera's angle towards the object) changes. To perform a multi-angle attack, we propose Adversarial Texture (AdvTexture). AdvTexture can cover clothes with arbitrary shapes so that people wearing such clothes can hide from person detectors from different viewing angles. We propose a generative method, named Toroidal-Cropping-based Expandable Generative Attack (TC-EGA), to craft AdvTexture with repetitive structures. We printed several pieces of cloth with AdvTexure and then made T-shirts, skirts, and dresses in the physical world. Experiments showed that these clothes could fool person detectors in the physical world.

17.L2CS-Net: Fine-Grained Gaze Estimation in Unconstrained Environments ⬇️

Human gaze is a crucial cue used in various applications such as human-robot interaction and virtual reality. Recently, convolution neural network (CNN) approaches have made notable progress in predicting gaze direction. However, estimating gaze in-the-wild is still a challenging problem due to the uniqueness of eye appearance, lightning conditions, and the diversity of head pose and gaze directions. In this paper, we propose a robust CNN-based model for predicting gaze in unconstrained settings. We propose to regress each gaze angle separately to improve the per-angel prediction accuracy, which will enhance the overall gaze performance. In addition, we use two identical losses, one for each angle, to improve network learning and increase its generalization. We evaluate our model with two popular datasets collected with unconstrained settings. Our proposed model achieves state-of-the-art accuracy of 3.92° and 10.41° on MPIIGaze and Gaze360 datasets, respectively. We make our code open source at this https URL.

18.Continuous Self-Localization on Aerial Images Using Visual and Lidar Sensors ⬇️

This paper proposes a novel method for geo-tracking, i.e. continuous metric self-localization in outdoor environments by registering a vehicle's sensor information with aerial imagery of an unseen target region. Geo-tracking methods offer the potential to supplant noisy signals from global navigation satellite systems (GNSS) and expensive and hard to maintain prior maps that are typically used for this purpose. The proposed geo-tracking method aligns data from on-board cameras and lidar sensors with geo-registered orthophotos to continuously localize a vehicle. We train a model in a metric learning setting to extract visual features from ground and aerial images. The ground features are projected into a top-down perspective via the lidar points and are matched with the aerial features to determine the relative pose between vehicle and orthophoto.
Our method is the first to utilize on-board cameras in an end-to-end differentiable model for metric self-localization on unseen orthophotos. It exhibits strong generalization, is robust to changes in the environment and requires only geo-poses as ground truth. We evaluate our approach on the KITTI-360 dataset and achieve a mean absolute position error (APE) of 0.94m. We further compare with previous approaches on the KITTI odometry dataset and achieve state-of-the-art results on the geo-tracking task.

19.Open Set Domain Adaptation By Novel Class Discovery ⬇️

In Open Set Domain Adaptation (OSDA), large amounts of target samples are drawn from the implicit categories that never appear in the source domain. Due to the lack of their specific belonging, existing methods indiscriminately regard them as a single class unknown. We challenge this broadly-adopted practice that may arouse unexpected detrimental effects because the decision boundaries between the implicit categories have been fully ignored. Instead, we propose Self-supervised Class-Discovering Adapter (SCDA) that attempts to achieve OSDA by gradually discovering those implicit classes, then incorporating them to restructure the classifier and update the domain-adaptive features iteratively. SCDA performs two alternate steps to achieve implicit class discovery and self-supervised OSDA, respectively. By jointly optimizing for two tasks, SCDA achieves the state-of-the-art in OSDA and shows a competitive performance to unearth the implicit target classes.

20.Least Square Estimation Network for Depth Completion ⬇️

Depth completion is a fundamental task in computer vision and robotics research. Many previous works complete the dense depth map with neural networks directly but most of them are non-interpretable and can not generalize to different situations well. In this paper, we propose an effective image representation method for depth completion tasks. The input of our system is a monocular camera frame and the synchronous sparse depth map. The output of our system is a dense per-pixel depth map of the frame. First we use a neural network to transform each pixel into a feature vector, which we call base functions. Then we pick out the known pixels' base functions and their depth values. We use a linear least square algorithm to fit the base functions and the depth values. Then we get the weights estimated from the least square algorithm. Finally, we apply the weights to the whole image and predict the final depth map. Our method is interpretable so it can generalize well. Experiments show that our results beat the state-of-the-art on NYU-Depth-V2 dataset both in accuracy and runtime. Moreover, experiments show that our method can generalize well on different numbers of sparse points and different datasets.

21.Comprehensive Review of Deep Learning-Based 3D Point Clouds Completion Processing and Analysis ⬇️

Point cloud completion is a generation and estimation issue derived from the partial point clouds, which plays a vital role in the applications in 3D computer vision. The progress of deep learning (DL) has impressively improved the capability and robustness of point cloud completion. However, the quality of completed point clouds is still needed to be further enhanced to meet the practical utilization. Therefore, this work aims to conduct a comprehensive survey on various methods, including point-based, convolution-based, graph-based, and generative model-based approaches, etc. And this survey summarizes the comparisons among these methods to provoke further research insights. Besides, this review sums up the commonly used datasets and illustrates the applications of point cloud completion. Eventually, we also discussed possible research trends in this promptly expanding field.

22.Interpretable part-whole hierarchies and conceptual-semantic relationships in neural networks ⬇️

Deep neural networks achieve outstanding results in a large variety of tasks, often outperforming human experts. However, a known limitation of current neural architectures is the poor accessibility to understand and interpret the network response to a given input. This is directly related to the huge number of variables and the associated non-linearities of neural models, which are often used as black boxes. When it comes to critical applications as autonomous driving, security and safety, medicine and health, the lack of interpretability of the network behavior tends to induce skepticism and limited trustworthiness, despite the accurate performance of such systems in the given task. Furthermore, a single metric, such as the classification accuracy, provides a non-exhaustive evaluation of most real-world scenarios. In this paper, we want to make a step forward towards interpretability in neural networks, providing new tools to interpret their behavior. We present Agglomerator, a framework capable of providing a representation of part-whole hierarchies from visual cues and organizing the input distribution matching the conceptual-semantic hierarchical structure between classes. We evaluate our method on common datasets, such as SmallNORB, MNIST, FashionMNIST, CIFAR-10, and CIFAR-100, providing a more interpretable model than other state-of-the-art approaches.

23.Dynamic MLP for Fine-Grained Image Classification by Leveraging Geographical and Temporal Information ⬇️

Fine-grained image classification is a challenging computer vision task where various species share similar visual appearances, resulting in misclassification if merely based on visual clues. Therefore, it is helpful to leverage additional information, e.g., the locations and dates for data shooting, which can be easily accessible but rarely exploited. In this paper, we first demonstrate that existing multimodal methods fuse multiple features only on a single dimension, which essentially has insufficient help in feature discrimination. To fully explore the potential of multimodal information, we propose a dynamic MLP on top of the image representation, which interacts with multimodal features at a higher and broader dimension. The dynamic MLP is an efficient structure parameterized by the learned embeddings of variable locations and dates. It can be regarded as an adaptive nonlinear projection for generating more discriminative image representations in visual tasks. To our best knowledge, it is the first attempt to explore the idea of dynamic networks to exploit multimodal information in fine-grained image classification tasks. Extensive experiments demonstrate the effectiveness of our method. The t-SNE algorithm visually indicates that our technique improves the recognizability of image representations that are visually similar but with different categories. Furthermore, among published works across multiple fine-grained datasets, dynamic MLP consistently achieves SOTA results this https URL and takes third place in the iNaturalist challenge at FGVC8 this https URL. Code is available at this https URL

24.Comparison of Spatio-Temporal Models for Human Motion and Pose Forecasting in Face-to-Face Interaction Scenarios ⬇️

Human behavior forecasting during human-human interactions is of utmost importance to provide robotic or virtual agents with social intelligence. This problem is especially challenging for scenarios that are highly driven by interpersonal dynamics. In this work, we present the first systematic comparison of state-of-the-art approaches for behavior forecasting. To do so, we leverage whole-body annotations (face, body, and hands) from the very recently released UDIVA v0.5, which features face-to-face dyadic interactions. Our best attention-based approaches achieve state-of-the-art performance in UDIVA v0.5. We show that by autoregressively predicting the future with methods trained for the short-term future (<400ms), we outperform the baselines even for a considerably longer-term future (up to 2s). We also show that this finding holds when highly noisy annotations are used, which opens new horizons towards the use of weakly-supervised learning. Combined with large-scale datasets, this may help boost the advances in this field.

25.Semantic Segmentation in Art Paintings ⬇️

Semantic segmentation is a difficult task even when trained in a supervised manner on photographs. In this paper, we tackle the problem of semantic segmentation of artistic paintings, an even more challenging task because of a much larger diversity in colors, textures, and shapes and because there are no ground truth annotations available for segmentation. We propose an unsupervised method for semantic segmentation of paintings using domain adaptation. Our approach creates a training set of pseudo-paintings in specific artistic styles by using style-transfer on the PASCAL VOC 2012 dataset, and then applies domain confusion between PASCAL VOC 2012 and real paintings. These two steps build on a new dataset we gathered called DRAM (Diverse Realism in Art Movements) composed of figurative art paintings from four movements, which are highly diverse in pattern, color, and geometry. To segment new paintings, we present a composite multi-domain adaptation method that trains on each sub-domain separately and composes their solutions during inference time. Our method provides better segmentation results not only on the specific artistic movements of DRAM, but also on other, unseen ones. We compare our approach to alternative methods and show applications of semantic segmentation in art paintings. The code and models for our approach are publicly available at: this https URL.

26.Signature and Log-signature for the Study of Empirical Distributions Generated with GANs ⬇️

In this paper, we develop a new and systematic method to explore and analyze samples taken by NASA Perseverance on the surface of the planet Mars. A novel in this context PCA adaptive t-SNE is proposed, as well as the introduction of statistical measures to study the goodness of fit of the sample distribution. We go beyond visualization by generating synthetic imagery using Stylegan2-ADA that resemble the original terrain distribution. We also conduct synthetic image generation using the recently introduced Scored-based Generative Modeling. We bring forward the use of the recently developed Signature Transform as a way to measure the similarity between image distributions and provide detailed acquaintance and extensive evaluations. We are the first to pioneer RMSE and MAE Signature and log-signature as an alternative to measure GAN convergence. Insights on state-of-the-art instance segmentation of the samples by the use of a model DeepLabv3 are also given.

27.Unpaired Image Captioning by Image-level Weakly-Supervised Visual Concept Recognition ⬇️

The goal of unpaired image captioning (UIC) is to describe images without using image-caption pairs in the training phase. Although challenging, we except the task can be accomplished by leveraging a training set of images aligned with visual concepts. Most existing studies use off-the-shelf algorithms to obtain the visual concepts because the Bounding Box (BBox) labels or relationship-triplet labels used for the training are expensive to acquire. In order to resolve the problem in expensive annotations, we propose a novel approach to achieve cost-effective UIC. Specifically, we adopt image-level labels for the optimization of the UIC model in a weakly-supervised manner. For each image, we assume that only the image-level labels are available without specific locations and numbers. The image-level labels are utilized to train a weakly-supervised object recognition model to extract object information (e.g., instance) in an image, and the extracted instances are adopted to infer the relationships among different objects based on an enhanced graph neural network (GNN). The proposed approach achieves comparable or even better performance compared with previous methods without the expensive cost of annotations. Furthermore, we design an unrecognized object (UnO) loss combined with a visual concept reward to improve the alignment of the inferred object and relationship information with the images. It can effectively alleviate the issue encountered by existing UIC models about generating sentences with nonexistent objects. To the best of our knowledge, this is the first attempt to solve the problem of Weakly-Supervised visual concept recognition for UIC (WS-UIC) based only on image-level labels. Extensive experiments have been carried out to demonstrate that the proposed WS-UIC model achieves inspiring results on the COCO dataset while significantly reducing the cost of labeling.

28.Knowledge Amalgamation for Object Detection with Transformers ⬇️

Knowledge amalgamation (KA) is a novel deep model reusing task aiming to transfer knowledge from several well-trained teachers to a multi-talented and compact student. Currently, most of these approaches are tailored for convolutional neural networks (CNNs). However, there is a tendency that transformers, with a completely different architecture, are starting to challenge the domination of CNNs in many computer vision tasks. Nevertheless, directly applying the previous KA methods to transformers leads to severe performance degradation. In this work, we explore a more effective KA scheme for transformer-based object detection models. Specifically, considering the architecture characteristics of transformers, we propose to dissolve the KA into two aspects: sequence-level amalgamation (SA) and task-level amalgamation (TA). In particular, a hint is generated within the sequence-level amalgamation by concatenating teacher sequences instead of redundantly aggregating them to a fixed-size one as previous KA works. Besides, the student learns heterogeneous detection tasks through soft targets with efficiency in the task-level amalgamation. Extensive experiments on PASCAL VOC and COCO have unfolded that the sequence-level amalgamation significantly boosts the performance of students, while the previous methods impair the students. Moreover, the transformer-based students excel in learning amalgamated knowledge, as they have mastered heterogeneous detection tasks rapidly and achieved superior or at least comparable performance to those of the teachers in their specializations.

29.CROON: Automatic Multi-LiDAR Calibration and Refinement Method in Road Scene ⬇️

Sensor-based environmental perception is a crucial part of the autonomous driving system. In order to get an excellent perception of the surrounding environment, an intelligent system would configure multiple LiDARs (3D Light Detection and Ranging) to cover the distant and near space of the car. The precision of perception relies on the quality of sensor calibration. This research aims at developing an accurate, automatic, and robust calibration strategy for multiple LiDAR systems in the general road scene. We thus propose CROON (automatiC multi-LiDAR CalibratiOn and Refinement method in rOad sceNe), a two-stage method including rough and refinement calibration. The first stage can calibrate the sensor from an arbitrary initial pose, and the second stage is able to precisely calibrate the sensor iteratively. Specifically, CROON utilize the nature characteristics of road scene so that it is independent and easy to apply in large-scale conditions. Experimental results on real-world and simulated data sets demonstrate the reliability and accuracy of our method. All the related data sets and codes are open-sourced on the Github website this https URL.

30.Dynamic Template Selection Through Change Detection for Adaptive Siamese Tracking ⬇️

Deep Siamese trackers have recently gained much attention in recent years since they can track visual objects at high speeds. Additionally, adaptive tracking methods, where target samples collected by the tracker are employed for online learning, have achieved state-of-the-art accuracy. However, single object tracking (SOT) remains a challenging task in real-world application due to changes and deformations in a target object's appearance. Learning on all the collected samples may lead to catastrophic forgetting, and thereby corrupt the tracking model.
In this paper, SOT is formulated as an online incremental learning problem. A new method is proposed for dynamic sample selection and memory replay, preventing template corruption. In particular, we propose a change detection mechanism to detect gradual changes in object appearance and select the corresponding samples for online adaption. In addition, an entropy-based sample selection strategy is introduced to maintain a diversified auxiliary buffer for memory replay. Our proposed method can be integrated into any object tracking algorithm that leverages online learning for model adaptation.
Extensive experiments conducted on the OTB-100, LaSOT, UAV123, and TrackingNet datasets highlight the cost-effectiveness of our method, along with the contribution of its key components. Results indicate that integrating our proposed method into state-of-art adaptive Siamese trackers can increase the potential benefits of a template update strategy, and significantly improve performance.

31.SingleSketch2Mesh : Generating 3D Mesh model from Sketch ⬇️

Sketching is an important activity in any design process. Designers and stakeholders share their ideas through hand-drawn sketches. These sketches are further used to create 3D models. Current methods to generate 3D models from sketches are either manual or tightly coupled with 3D modeling platforms. Therefore, it requires users to have an experience of sketching on such platform. Moreover, most of the existing approaches are based on geometric manipulation and thus cannot be generalized. We propose a novel AI based ensemble approach, SingleSketch2Mesh, for generating 3D models from hand-drawn sketches. Our approach is based on Generative Networks and Encoder-Decoder Architecture to generate 3D mesh model from a hand-drawn sketch. We evaluate our solution with existing solutions. Our approach outperforms existing approaches on both - quantitative and qualitative evaluation criteria.

32.On the Construction of Distribution-Free Prediction Intervals for an Image Regression Problem in Semiconductor Manufacturing ⬇️

The high-volume manufacturing of the next generation of semiconductor devices requires advances in measurement signal analysis. Many in the semiconductor manufacturing community have reservations about the adoption of deep learning; they instead prefer other model-based approaches for some image regression problems, and according to the 2021 IEEE International Roadmap for Devices and Systems (IRDS) report on Metrology a SEMI standardization committee may endorse this philosophy. The semiconductor manufacturing community does, however, communicate a need for state-of-the-art statistical analyses to reduce measurement uncertainty. Prediction intervals which characterize the reliability of the predictive performance of regression models can impact decisions, build trust in machine learning, and be applied to other regression models. However, we are not aware of effective and sufficiently simple distribution-free approaches that offer valid coverage for important classes of image data, so we consider the distribution-free conformal prediction and conformalized quantile regression framework.The image regression problem that is the focus of this paper pertains to line edge roughness (LER) estimation from noisy scanning electron microscopy images. LER affects semiconductor device performance and reliability as well as the yield of the manufacturing process; the 2021 IRDS emphasizes the crucial importance of LER by devoting a white paper to it in addition to mentioning or discussing it in the reports of multiple international focus teams. It is not immediately apparent how to effectively use normalized conformal prediction and quantile regression for LER estimation. The modeling techniques we apply appear to be novel for finding distribution-free prediction intervals for image data and will be presented at the 2022 SEMI Advanced Semiconductor Manufacturing Conference.

33.End-to-end video instance segmentation via spatial-temporal graph neural networks ⬇️

Video instance segmentation is a challenging task that extends image instance segmentation to the video domain. Existing methods either rely only on single-frame information for the detection and segmentation subproblems or handle tracking as a separate post-processing step, which limit their capability to fully leverage and share useful spatial-temporal information for all the subproblems. In this paper, we propose a novel graph-neural-network (GNN) based method to handle the aforementioned limitation. Specifically, graph nodes representing instance features are used for detection and segmentation while graph edges representing instance relations are used for tracking. Both inter and intra-frame information is effectively propagated and shared via graph updates and all the subproblems (i.e. detection, segmentation and tracking) are jointly optimized in an unified framework. The performance of our method shows great improvement on the YoutubeVIS validation dataset compared to existing methods and achieves 35.2% AP with a ResNet-50 backbone, operating at 22 FPS. Code is available at this http URL .

34.MSDN: Mutually Semantic Distillation Network for Zero-Shot Learning ⬇️

The key challenge of zero-shot learning (ZSL) is how to infer the latent semantic knowledge between visual and attribute features on seen classes, and thus achieving a desirable knowledge transfer to unseen classes. Prior works either simply align the global features of an image with its associated class semantic vector or utilize unidirectional attention to learn the limited latent semantic representations, which could not effectively discover the intrinsic semantic knowledge e.g., attribute semantics) between visual and attribute features. To solve the above dilemma, we propose a Mutually Semantic Distillation Network (MSDN), which progressively distills the intrinsic semantic representations between visual and attribute features for ZSL. MSDN incorporates an attribute$\rightarrow$visual attention sub-net that learns attribute-based visual features, and a visual$\rightarrow$attribute attention sub-net that learns visual-based attribute features. By further introducing a semantic distillation loss, the two mutual attention sub-nets are capable of learning collaboratively and teaching each other throughout the training process. The proposed MSDN yields significant improvements over the strong baselines, leading to new state-of-the-art performances on three popular challenging benchmarks, i.e., CUB, SUN, and AWA2. Our codes have been available at: \url{this https URL}.

35.Protecting Facial Privacy: Generating Adversarial Identity Masks via Style-robust Makeup Transfer ⬇️

While deep face recognition (FR) systems have shown amazing performance in identification and verification, they also arouse privacy concerns for their excessive surveillance on users, especially for public face images widely spread on social networks. Recently, some studies adopt adversarial examples to protect photos from being identified by unauthorized face recognition systems. However, existing methods of generating adversarial face images suffer from many limitations, such as awkward visual, white-box setting, weak transferability, making them difficult to be applied to protect face privacy in reality. In this paper, we propose adversarial makeup transfer GAN (AMT-GAN), a novel face protection method aiming at constructing adversarial face images that preserve stronger black-box transferability and better visual quality simultaneously. AMT-GAN leverages generative adversarial networks (GAN) to synthesize adversarial face images with makeup transferred from reference images. In particular, we introduce a new regularization module along with a joint training strategy to reconcile the conflicts between the adversarial noises and the cycle consistence loss in makeup transfer, achieving a desirable balance between the attack strength and visual changes. Extensive experiments verify that compared with state of the arts, AMT-GAN can not only preserve a comfortable visual quality, but also achieve a higher attack success rate over commercial FR APIs, including Face++, Aliyun, and Microsoft.

36.Behavior Recognition Based on the Integration of Multigranular Motion Features ⬇️

The recognition of behaviors in videos usually requires a combinatorial analysis of the spatial information about objects and their dynamic action information in the temporal dimension. Specifically, behavior recognition may even rely more on the modeling of temporal information containing short-range and long-range motions; this contrasts with computer vision tasks involving images that focus on the understanding of spatial information. However, current solutions fail to jointly and comprehensively analyze short-range motion between adjacent frames and long-range temporal aggregations at large scales in videos. In this paper, we propose a novel behavior recognition method based on the integration of multigranular (IMG) motion features. In particular, we achieve reliable motion information modeling through the synergy of a channel attention-based short-term motion feature enhancement module (CMEM) and a cascaded long-term motion feature integration module (CLIM). We evaluate our model on several action recognition benchmarks such as HMDB51, Something-Something and UCF101. The experimental results demonstrate that our approach outperforms the previous state-of-the-art methods, which confirms its effectiveness and efficiency.

37.CPPF: Towards Robust Category-Level 9D Pose Estimation in the Wild ⬇️

In this paper, we tackle the problem of category-level 9D pose estimation in the wild, given a single RGB-D frame. Using supervised data of real-world 9D poses is tedious and erroneous, and also fails to generalize to unseen scenarios. Besides, category-level pose estimation requires a method to be able to generalize to unseen objects at test time, which is also challenging. Drawing inspirations from traditional point pair features (PPFs), in this paper, we design a novel Category-level PPF (CPPF) voting method to achieve accurate, robust and generalizable 9D pose estimation in the wild. To obtain robust pose estimation, we sample numerous point pairs on an object, and for each pair our model predicts necessary SE(3)-invariant voting statistics on object centers, orientations and scales. A novel coarse-to-fine voting algorithm is proposed to eliminate noisy point pair samples and generate final predictions from the population. To get rid of false positives in the orientation voting process, an auxiliary binary disambiguating classification task is introduced for each sampled point pair. In order to detect objects in the wild, we carefully design our sim-to-real pipeline by training on synthetic point clouds only, unless objects have ambiguous poses in geometry. Under this circumstance, color information is leveraged to disambiguate these poses. Results on standard benchmarks show that our method is on par with current state of the arts with real-world training data. Extensive experiments further show that our method is robust to noise and gives promising results under extremely challenging scenarios. Our code is available on this https URL.

38.HAR-GCNN: Deep Graph CNNs for Human Activity Recognition From Highly Unlabeled Mobile Sensor Data ⬇️

The problem of human activity recognition from mobile sensor data applies to multiple domains, such as health monitoring, personal fitness, daily life logging, and senior care. A critical challenge for training human activity recognition models is data quality. Acquiring balanced datasets containing accurate activity labels requires humans to correctly annotate and potentially interfere with the subjects' normal activities in real-time. Despite the likelihood of incorrect annotation or lack thereof, there is often an inherent chronology to human behavior. For example, we take a shower after we exercise. This implicit chronology can be used to learn unknown labels and classify future activities. In this work, we propose HAR-GCCN, a deep graph CNN model that leverages the correlation between chronologically adjacent sensor measurements to predict the correct labels for unclassified activities that have at least one activity label. We propose a new training strategy enforcing that the model predicts the missing activity labels by leveraging the known ones. HAR-GCCN shows superior performance relative to previously used baseline methods, improving classification accuracy by about 25% and up to 68% on different datasets. Code is available at \url{this https URL}.

39.GlideNet: Global, Local and Intrinsic based Dense Embedding NETwork for Multi-category Attributes Prediction ⬇️

Attaching attributes (such as color, shape, state, action) to object categories is an important computer vision problem. Attribute prediction has seen exciting recent progress and is often formulated as a multi-label classification problem. Yet significant challenges remain in: 1) predicting diverse attributes over multiple categories, 2) modeling attributes-category dependency, 3) capturing both global and local scene context, and 4) predicting attributes of objects with low pixel-count. To address these issues, we propose a novel multi-category attribute prediction deep architecture named GlideNet, which contains three distinct feature extractors. A global feature extractor recognizes what objects are present in a scene, whereas a local one focuses on the area surrounding the object of interest. Meanwhile, an intrinsic feature extractor uses an extension of standard convolution dubbed Informed Convolution to retrieve features of objects with low pixel-count. GlideNet uses gating mechanisms with binary masks and its self-learned category embedding to combine the dense embeddings. Collectively, the Global-Local-Intrinsic blocks comprehend the scene's global context while attending to the characteristics of the local object of interest. Finally, using the combined features, an interpreter predicts the attributes, and the length of the output is determined by the category, thereby removing unnecessary attributes. GlideNet can achieve compelling results on two recent and challenging datasets -- VAW and CAR -- for large-scale attribute prediction. For instance, it obtains more than 5% gain over state of the art in the mean recall (mR) metric. GlideNet's advantages are especially apparent when predicting attributes of objects with low pixel counts as well as attributes that demand global context understanding. Finally, we show that GlideNet excels in training starved real-world scenarios.

40.Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimation ⬇️

Best-of-N (BoN) Average Displacement Error (ADE)/ Final Displacement Error (FDE) is the most used metric for evaluating trajectory prediction models. Yet, the BoN does not quantify the whole generated samples, resulting in an incomplete view of the model's prediction quality and performance. We propose a new metric, Average Mahalanobis Distance (AMD) to tackle this issue. AMD is a metric that quantifies how close the whole generated samples are to the ground truth. We also introduce the Average Maximum Eigenvalue (AMV) metric that quantifies the overall spread of the predictions. Our metrics are validated empirically by showing that the ADE/FDE is not sensitive to distribution shifts, giving a biased sense of accuracy, unlike the AMD/AMV metrics. We introduce the usage of Implicit Maximum Likelihood Estimation (IMLE) as a replacement for traditional generative models to train our model, Social-Implicit. IMLE training mechanism aligns with AMD/AMV objective of predicting trajectories that are close to the ground truth with a tight spread. Social-Implicit is a memory efficient deep model with only 5.8K parameters that runs in real time of about 580Hz and achieves competitive results. Interactive demo of the problem can be seen here \url{this https URL}. Code is available at \url{this https URL}.

41.Highly Accurate Dichotomous Image Segmentation ⬇️

We present a systematic study on a new task called dichotomous image segmentation (DIS), which aims to segment highly accurate objects from natural images. To this end, we collected the first large-scale dataset, called DIS5K, which contains 5,470 high-resolution (e.g., 2K, 4K or larger) images covering camouflaged, salient, or meticulous objects in various backgrounds. All images are annotated with extremely fine-grained labels. In addition, we introduce a simple intermediate supervision baseline (IS-Net) using both feature-level and mask-level guidance for DIS model training. Without tricks, IS-Net outperforms various cutting-edge baselines on the proposed DIS5K, making it a general self-learned supervision network that can help facilitate future research in DIS. Further, we design a new metric called human correction efforts (HCE) which approximates the number of mouse clicking operations required to correct the false positives and false negatives. HCE is utilized to measure the gap between models and real-world applications and thus can complement existing metrics. Finally, we conduct the largest-scale benchmark, evaluating 16 representative segmentation models, providing a more insightful discussion regarding object complexities, and showing several potential applications (e.g., background removal, art design, 3D reconstruction). Hoping these efforts can open up promising directions for both academic and industries. We will release our DIS5Kdataset, IS-Net baseline, HCE metric, and the complete benchmark results.

42.Learnable Irrelevant Modality Dropout for Multimodal Action Recognition on Modality-Specific Annotated Videos ⬇️

With the assumption that a video dataset is multimodality annotated in which auditory and visual modalities both are labeled or class-relevant, current multimodal methods apply modality fusion or cross-modality attention. However, effectively leveraging the audio modality in vision-specific annotated videos for action recognition is of particular challenge. To tackle this challenge, we propose a novel audio-visual framework that effectively leverages the audio modality in any solely vision-specific annotated dataset. We adopt the language models (e.g., BERT) to build a semantic audio-video label dictionary (SAVLD) that maps each video label to its most K-relevant audio labels in which SAVLD serves as a bridge between audio and video datasets. Then, SAVLD along with a pretrained audio multi-label model are used to estimate the audio-visual modality relevance during the training phase. Accordingly, a novel learnable irrelevant modality dropout (IMD) is proposed to completely drop out the irrelevant audio modality and fuse only the relevant modalities. Moreover, we present a new two-stream video Transformer for efficiently modeling the visual modalities. Results on several vision-specific annotated datasets including Kinetics400 and UCF-101 validated our framework as it outperforms most relevant action recognition methods.

43.Semantic-Aware Latent Space Exploration for Face Image Restoration ⬇️

For image restoration, most existing deep learning based methods tend to overfit the training data leading to bad results when encountering unseen degradations out of the assumptions for training. To improve the robustness, generative adversarial network (GAN) prior based methods have been proposed, revealing a promising capability to restore photo-realistic and high-quality results. But these methods suffer from semantic confusion, especially on semantically significant images such as face images. In this paper, we propose a semantic-aware latent space exploration method for image restoration (SAIR). By explicitly modeling referenced semantics information, SAIR can consistently restore severely degraded images not only to high-resolution highly-realistic looks but also to correct semantics. Quantitative and qualitative experiments collectively demonstrate the effectiveness of the proposed SAIR. Our code can be found in this https URL.

44.Modeling Coreference Relations in Visual Dialog ⬇️

Visual dialog is a vision-language task where an agent needs to answer a series of questions grounded in an image based on the understanding of the dialog history and the image. The occurrences of coreference relations in the dialog makes it a more challenging task than visual question-answering. Most previous works have focused on learning better multi-modal representations or on exploring different ways of fusing visual and language features, while the coreferences in the dialog are mainly ignored. In this paper, based on linguistic knowledge and discourse features of human dialog we propose two soft constraints that can improve the model's ability of resolving coreferences in dialog in an unsupervised way. Experimental results on the VisDial v1.0 dataset shows that our model, which integrates two novel and linguistically inspired soft constraints in a deep transformer neural architecture, obtains new state-of-the-art performance in terms of recall at 1 and other evaluation metrics compared to current existing models and this without pretraining on other vision-language datasets. Our qualitative results also demonstrate the effectiveness of the method that we propose.

45.Dynamic Key-value Memory Enhanced Multi-step Graph Reasoning for Knowledge-based Visual Question Answering ⬇️

Knowledge-based visual question answering (VQA) is a vision-language task that requires an agent to correctly answer image-related questions using knowledge that is not presented in the given image. It is not only a more challenging task than regular VQA but also a vital step towards building a general VQA system. Most existing knowledge-based VQA systems process knowledge and image information similarly and ignore the fact that the knowledge base (KB) contains complete information about a triplet, while the extracted image information might be incomplete as the relations between two objects are missing or wrongly detected. In this paper, we propose a novel model named dynamic knowledge memory enhanced multi-step graph reasoning (DMMGR), which performs explicit and implicit reasoning over a key-value knowledge memory module and a spatial-aware image graph, respectively. Specifically, the memory module learns a dynamic knowledge representation and generates a knowledge-aware question representation at each reasoning step. Then, this representation is used to guide a graph attention operator over the spatial-aware image graph. Our model achieves new state-of-the-art accuracy on the KRVQR and FVQA datasets. We also conduct ablation experiments to prove the effectiveness of each component of the proposed model.

46.Exploring Optical-Flow-Guided Motion and Detection-Based Appearance for Temporal Sentence Grounding ⬇️

Temporal sentence grounding aims to localize a target segment in an untrimmed video semantically according to a given sentence query. Most previous works focus on learning frame-level features of each whole frame in the entire video, and directly match them with the textual information. Such frame-level feature extraction leads to the obstacles of these methods in distinguishing ambiguous video frames with complicated contents and subtle appearance differences, thus limiting their performance. In order to differentiate fine-grained appearance similarities among consecutive frames, some state-of-the-art methods additionally employ a detection model like Faster R-CNN to obtain detailed object-level features in each frame for filtering out the redundant background contents. However, these methods suffer from missing motion analysis since the object detection module in Faster R-CNN lacks temporal modeling. To alleviate the above limitations, in this paper, we propose a novel Motion- and Appearance-guided 3D Semantic Reasoning Network (MA3SRN), which incorporates optical-flow-guided motion-aware, detection-based appearance-aware, and 3D-aware object-level features to better reason the spatial-temporal object relations for accurately modelling the activity among consecutive frames. Specifically, we first develop three individual branches for motion, appearance, and 3D encoding separately to learn fine-grained motion-guided, appearance-guided, and 3D-aware object features, respectively. Then, both motion and appearance information from corresponding branches are associated to enhance the 3D-aware features for the final precise grounding. Extensive experiments on three challenging datasets (ActivityNet Caption, Charades-STA and TACoS) demonstrate that the proposed MA3SRN model achieves a new state-of-the-art.

47.Precise Point Spread Function Estimation ⬇️

Point spread function (PSF) plays a crucial role in many fields, such as shape from focus/defocus, depth estimation, and imaging process in fluorescence microscopy. However, the mathematical model of the defocus process is still unclear because several variables in the point spread function are hard to measure accurately, such as the f-number of cameras, the physical size of a pixel, the focus depth, etc. In this work, we develop a precise mathematical model of the camera's point spread function to describe the defocus process. We first derive the mathematical algorithm for the PSF and extract two parameters A and e. A is the composite of camera's f-number, pixel-size, output scale, and scaling factor of the circle of confusion; e is the deviation of the focus depth. We design a novel metric based on the defocus histogram to evaluate the difference between the simulated focused image and the actual focused image to obtain optimal A and e. We also construct a hardware system consisting of a focusing system and a structured light system to acquire the all-in-focus image, the focused image with corresponding focus depth, and the depth map in the same view. The three types of images, as a dataset, are used to obtain the precise PSF. Our experiments on standard planes and actual objects show that the proposed algorithm can accurately describe the defocus process. The accuracy of our algorithm is further proved by evaluating the difference among the actual focused images, the focused image generated by our algorithm, the focused image generated by others. The results show that the loss of our algorithm is 40% less than others on average. The dataset, code, and model are available on GitHub: this https URL precise-point-spread-function-estimation.

48.On Steering Multi-Annotations per Sample for Multi-Task Learning ⬇️

The study of multi-task learning has drawn great attention from the community. Despite the remarkable progress, the challenge of optimally learning different tasks simultaneously remains to be explored. Previous works attempt to modify the gradients from different tasks. Yet these methods give a subjective assumption of the relationship between tasks, and the modified gradient may be less accurate. In this paper, we introduce Stochastic Task Allocation~(STA), a mechanism that addresses this issue by a task allocation approach, in which each sample is randomly allocated a subset of tasks. For further progress, we propose Interleaved Stochastic Task Allocation~(ISTA) to iteratively allocate all tasks to each example during several consecutive iterations. We evaluate STA and ISTA on various datasets and applications: NYUv2, Cityscapes, and COCO for scene understanding and instance segmentation. Our experiments show both STA and ISTA outperform current state-of-the-art methods. The code will be available.

49.Detection of Parasitic Eggs from Microscopy Images and the emergence of a new dataset ⬇️

Automatic detection of parasitic eggs in microscopy images has the potential to increase the efficiency of human experts whilst also providing an objective assessment. The time saved by such a process would both help ensure a prompt treatment to patients, and off-load excessive work from experts' shoulders. Advances in deep learning inspired us to exploit successful architectures for detection, adapting them to tackle a different domain. We propose a framework that exploits two such state-of-the-art models. Specifically, we demonstrate results produced by both a Generative Adversarial Network (GAN) and Faster-RCNN, for image enhancement and object detection respectively, on microscopy images of varying quality. The use of these techniques yields encouraging results, though further improvements are still needed for certain egg types whose detection still proves challenging. As a result, a new dataset has been created and made publicly available, providing an even wider range of classes and variability.

50.Weakly Supervised Temporal Action Localization via Representative Snippet Knowledge Propagation ⬇️

Weakly supervised temporal action localization aims to localize temporal boundaries of actions and simultaneously identify their categories with only video-level category labels. Many existing methods seek to generate pseudo labels for bridging the discrepancy between classification and localization, but usually only make use of limited contextual information for pseudo label generation. To alleviate this problem, we propose a representative snippet summarization and propagation framework. Our method seeks to mine the representative snippets in each video for propagating information between video snippets to generate better pseudo labels. For each video, its own representative snippets and the representative snippets from a memory bank are propagated to update the input features in an intra- and inter-video manner. The pseudo labels are generated from the temporal class activation maps of the updated features to rectify the predictions of the main branch. Our method obtains superior performance in comparison to the existing methods on two benchmarks, THUMOS14 and ActivityNet1.3, achieving gains as high as 1.2% in terms of average mAP on THUMOS14.

51.PanFormer: a Transformer Based Model for Pan-sharpening ⬇️

Pan-sharpening aims at producing a high-resolution (HR) multi-spectral (MS) image from a low-resolution (LR) multi-spectral (MS) image and its corresponding panchromatic (PAN) image acquired by a same satellite. Inspired by a new fashion in recent deep learning community, we propose a novel Transformer based model for pan-sharpening. We explore the potential of Transformer in image feature extraction and fusion. Following the successful development of vision transformers, we design a two-stream network with the self-attention to extract the modality-specific features from the PAN and MS modalities and apply a cross-attention module to merge the spectral and spatial features. The pan-sharpened image is produced from the enhanced fused features. Extensive experiments on GaoFen-2 and WorldView-3 images demonstrate that our Transformer based model achieves impressive results and outperforms many existing CNN based methods, which shows the great potential of introducing Transformer to the pan-sharpening task. Codes are available at this https URL.

52.Exploring Dual-task Correlation for Pose Guided Person Image Generation ⬇️

Pose Guided Person Image Generation (PGPIG) is the task of transforming a person image from the source pose to a given target pose. Most of the existing methods only focus on the ill-posed source-to-target task and fail to capture reasonable texture mapping. To address this problem, we propose a novel Dual-task Pose Transformer Network (DPTN), which introduces an auxiliary task (i.e., source-to-source task) and exploits the dual-task correlation to promote the performance of PGPIG. The DPTN is of a Siamese structure, containing a source-to-source self-reconstruction branch, and a transformation branch for source-to-target generation. By sharing partial weights between them, the knowledge learned by the source-to-source task can effectively assist the source-to-target learning. Furthermore, we bridge the two branches with a proposed Pose Transformer Module (PTM) to adaptively explore the correlation between features from dual tasks. Such correlation can establish the fine-grained mapping of all the pixels between the sources and the targets, and promote the source texture transmission to enhance the details of the generated target images. Extensive experiments show that our DPTN outperforms state-of-the-arts in terms of both PSNR and LPIPS. In addition, our DPTN only contains 9.79 million parameters, which is significantly smaller than other approaches. Our code is available at: this https URL.

53.Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentation ⬇️

Weakly Supervised Semantic Segmentation (WSSS) based on image-level labels has attracted much attention due to low annotation costs. Existing methods often rely on Class Activation Mapping (CAM) that measures the correlation between image pixels and classifier weight. However, the classifier focuses only on the discriminative regions while ignoring other useful information in each image, resulting in incomplete localization maps. To address this issue, we propose a Self-supervised Image-specific Prototype Exploration (SIPE) that consists of an Image-specific Prototype Exploration (IPE) and a General-Specific Consistency (GSC) loss. Specifically, IPE tailors prototypes for every image to capture complete regions, formed our Image-Specific CAM (IS-CAM), which is realized by two sequential steps. In addition, GSC is proposed to construct the consistency of general CAM and our specific IS-CAM, which further optimizes the feature representation and empowers a self-correction ability of prototype exploration. Extensive experiments are conducted on PASCAL VOC 2012 and MS COCO 2014 segmentation benchmark and results show our SIPE achieves new state-of-the-art performance using only image-level labels. The code is available at this https URL.

54.A Robust Framework of Chromosome Straightening with ViT-Patch GAN ⬇️

Chromosomes exhibit non-rigid and non-articulated nature with varying degrees of curvature. Chromosome straightening is an essential step for subsequent karyotype construction, pathological diagnosis and cytogenetic map development. However, robust chromosome straightening remains challenging, due to the unavailability of training images, distorted chromosome details and shapes after straightening, as well as poor generalization capability. We propose a novel architecture, ViT-Patch GAN, consisting of a motion transformation generator and a Vision Transformer-based patch (ViT-Patch) discriminator. The generator learns the motion representation of chromosomes for straightening. With the help of the ViT-Patch discriminator, the straightened chromosomes retain more shape and banding pattern details. The proposed framework is trained on a small dataset and is able to straighten chromosome images with state-of-the-art performance for two large datasets.

55.Multi-class Token Transformer for Weakly Supervised Semantic Segmentation ⬇️

This paper proposes a new transformer-based framework to learn class-specific object localization maps as pseudo labels for weakly supervised semantic segmentation (WSSS). Inspired by the fact that the attended regions of the one-class token in the standard vision transformer can be leveraged to form a class-agnostic localization map, we investigate if the transformer model can also effectively capture class-specific attention for more discriminative object localization by learning multiple class tokens within the transformer. To this end, we propose a Multi-class Token Transformer, termed as MCTformer, which uses multiple class tokens to learn interactions between the class tokens and the patch tokens. The proposed MCTformer can successfully produce class-discriminative object localization maps from class-to-patch attentions corresponding to different class tokens. We also propose to use a patch-level pairwise affinity, which is extracted from the patch-to-patch transformer attention, to further refine the localization maps. Moreover, the proposed framework is shown to fully complement the Class Activation Mapping (CAM) method, leading to remarkably superior WSSS results on the PASCAL VOC and MS COCO datasets. These results underline the importance of the class token for WSSS.

56.Towards Self-Supervised Category-Level Object Pose and Size Estimation ⬇️

This work presents a self-supervised framework for category-level object pose and size estimation from a single depth image. Unlike previous works that rely on time-consuming and labor-intensive ground truth pose labels for supervision, we leverage the geometric consistency residing in point clouds of the same shape for self-supervision. Specifically, given a normalized category template mesh in the object-coordinate system and the partially observed object instance in the scene, our key idea is to apply differentiable shape deformation, registration, and rendering to enforce geometric consistency between the predicted and the observed scene object point cloud. We evaluate our approach on real-world datasets and find that our approach outperforms the simple traditional baseline by large margins while being competitive with some fully-supervised approaches.

57.Region Proposal Rectification Towards Robust Instance Segmentation of Biological Images ⬇️

Top-down instance segmentation framework has shown its superiority in object detection compared to the bottom-up framework. While it is efficient in addressing over-segmentation, top-down instance segmentation suffers from over-crop problem. However, a complete segmentation mask is crucial for biological image analysis as it delivers important morphological properties such as shapes and volumes. In this paper, we propose a region proposal rectification (RPR) module to address this challenging incomplete segmentation problem. In particular, we offer a progressive ROIAlign module to introduce neighbor information into a series of ROIs gradually. The ROI features are fed into an attentive feed-forward network (FFN) for proposal box regression. With additional neighbor information, the proposed RPR module shows significant improvement in correction of region proposal locations and thereby exhibits favorable instance segmentation performances on three biological image datasets compared to state-of-the-art baseline methods. Experimental results demonstrate that the proposed RPR module is effective in both anchor-based and anchor-free top-down instance segmentation approaches, suggesting the proposed method can be applied to general top-down instance segmentation of biological images.

58.Evaluation of Dirichlet Process Gaussian Mixtures for Segmentation on Noisy Hyperspectral Images ⬇️

Image segmentation is a fundamental step for the interpretation of Remote Sensing Images. Clustering or segmentation methods usually precede the classification task and are used as support tools for manual labeling. The most common algorithms, such as k-means, mean-shift, and MRS, require an extra manual step to find the scale parameter. The segmentation results are severely affected if the parameters are not correctly tuned and diverge from the optimal values. Additionally, the search for the optimal scale is a costly task, as it requires a comprehensive hyper-parameter search. This paper proposes and evaluates a method for segmentation of Hyperspectral Images using the Dirichlet Process Gaussian Mixture Model. Our model can self-regulate the parameters until it finds the optimal values of scale and the number of clusters in a given dataset. The results demonstrate the potential of our method to find objects in a Hyperspectral Image while bypassing the burden of manual search of the optimal parameters. In addition, our model also produces similar results on noisy datasets, while previous research usually required a pre-processing task for noise reduction and spectral smoothing.

59.Adversarial Dual-Student with Differentiable Spatial Warping for Semi-Supervised Semantic Segmentation ⬇️

A common challenge posed to robust semantic segmentation is the expensive data annotation cost. Existing semi-supervised solutions show great potential toward solving this problem. Their key idea is constructing consistency regularization with unsupervised data augmentation from unlabeled data for model training. The perturbations for unlabeled data enable the consistency training loss, which benefits semi-supervised semantic segmentation. However, these perturbations destroy image context and introduce unnatural boundaries, which is harmful for semantic segmentation. Besides, the widely adopted semi-supervised learning framework, i.e. mean-teacher, suffers performance limitation since the student model finally converges to the teacher model. In this paper, first of all, we propose a context friendly differentiable geometric warping to conduct unsupervised data augmentation; secondly, a novel adversarial dual-student framework is proposed to improve the Mean-Teacher from the following two aspects: (1) dual student models are learnt independently except for a stabilization constraint to encourage exploiting model diversities; (2) adversarial training scheme is applied to both students and the discriminators are resorted to distinguish reliable pseudo-label of unlabeled data for self-training. Effectiveness is validated via extensive experiments on PASCAL VOC2012 and Citescapes. Our solution significantly improves the performance and state-of-the-art results are achieved on both datasets. Remarkably, compared with fully supervision, our solution achieves comparable mIoU of 73.4% using only 12.5% annotated data on PASCAL VOC2012.

60.Don't Be So Dense: Sparse-to-Sparse GAN Training Without Sacrificing Performance ⬇️

Generative adversarial networks (GANs) have received an upsurging interest since being proposed due to the high quality of the generated data. While achieving increasingly impressive results, the resource demands associated with the large model size hinders the usage of GANs in resource-limited scenarios. For inference, the existing model compression techniques can reduce the model complexity with comparable performance. However, the training efficiency of GANs has less been explored due to the fragile training process of GANs. In this paper, we, for the first time, explore the possibility of directly training sparse GAN from scratch without involving any dense or pre-training steps. Even more unconventionally, our proposed method enables directly training sparse unbalanced GANs with an extremely sparse generator from scratch. Instead of training full GANs, we start with sparse GANs and dynamically explore the parameter space spanned over the generator throughout training. Such a sparse-to-sparse training procedure enhances the capacity of the highly sparse generator progressively while sticking to a fixed small parameter budget with appealing training and inference efficiency gains. Extensive experiments with modern GAN architectures validate the effectiveness of our method. Our sparsified GANs, trained from scratch in one single run, are able to outperform the ones learned by expensive iterative pruning and re-training. Perhaps most importantly, we find instead of inheriting parameters from expensive pre-trained GANs, directly training sparse GANs from scratch can be a much more efficient solution. For example, only training with a 80% sparse generator and a 70% sparse discriminator, our method can achieve even better performance than the dense BigGAN.

61.Towards Robust Part-aware Instance Segmentation for Industrial Bin Picking ⬇️

Industrial bin picking is a challenging task that requires accurate and robust segmentation of individual object instances. Particularly, industrial objects can have irregular shapes, that is, thin and concave, whereas in bin-picking scenarios, objects are often closely packed with strong occlusion. To address these challenges, we formulate a novel part-aware instance segmentation pipeline. The key idea is to decompose industrial objects into correlated approximate convex parts and enhance the object-level segmentation with part-level segmentation. We design a part-aware network to predict part masks and part-to-part offsets, followed by a part aggregation module to assemble the recognized parts into instances. To guide the network learning, we also propose an automatic label decoupling scheme to generate ground-truth part-level labels from instance-level labels. Finally, we contribute the first instance segmentation dataset, which contains a variety of industrial objects that are thin and have non-trivial shapes. Extensive experimental results on various industrial objects demonstrate that our method can achieve the best segmentation results compared with the state-of-the-art approaches.

62.Bridging the Gap Between Learning in Discrete and Continuous Environments for Vision-and-Language Navigation ⬇️

Most existing works in vision-and-language navigation (VLN) focus on either discrete or continuous environments, training agents that cannot generalize across the two. The fundamental difference between the two setups is that discrete navigation assumes prior knowledge of the connectivity graph of the environment, so that the agent can effectively transfer the problem of navigation with low-level controls to jumping from node to node with high-level actions by grounding to an image of a navigable direction. To bridge the discrete-to-continuous gap, we propose a predictor to generate a set of candidate waypoints during navigation, so that agents designed with high-level actions can be transferred to and trained in continuous environments. We refine the connectivity graph of Matterport3D to fit the continuous Habitat-Matterport3D, and train the waypoints predictor with the refined graphs to produce accessible waypoints at each time step. Moreover, we demonstrate that the predicted waypoints can be augmented during training to diversify the views and paths, and therefore enhance agent's generalization ability. Through extensive experiments we show that agents navigating in continuous environments with predicted waypoints perform significantly better than agents using low-level actions, which reduces the absolute discrete-to-continuous gap by 11.76% Success Weighted by Path Length (SPL) for the Cross-Modal Matching Agent and 18.24% SPL for the Recurrent VLN-BERT. Our agents, trained with a simple imitation learning objective, outperform previous methods by a large margin, achieving new state-of-the-art results on the testing environments of the R2R-CE and the RxR-CE datasets.

63.MetaFormer: A Unified Meta Framework for Fine-Grained Recognition ⬇️

Fine-Grained Visual Classification(FGVC) is the task that requires recognizing the objects belonging to multiple subordinate categories of a super-category. Recent state-of-the-art methods usually design sophisticated learning pipelines to tackle this task. However, visual information alone is often not sufficient to accurately differentiate between fine-grained visual categories. Nowadays, the meta-information (e.g., spatio-temporal prior, attribute, and text description) usually appears along with the images. This inspires us to ask the question: Is it possible to use a unified and simple framework to utilize various meta-information to assist in fine-grained identification? To answer this problem, we explore a unified and strong meta-framework(MetaFormer) for fine-grained visual classification. In practice, MetaFormer provides a simple yet effective approach to address the joint learning of vision and various meta-information. Moreover, MetaFormer also provides a strong baseline for FGVC without bells and whistles. Extensive experiments demonstrate that MetaFormer can effectively use various meta-information to improve the performance of fine-grained recognition. In a fair comparison, MetaFormer can outperform the current SotA approaches with only vision information on the iNaturalist2017 and iNaturalist2018 datasets. Adding meta-information, MetaFormer can exceed the current SotA approaches by 5.9% and 5.3%, respectively. Moreover, MetaFormer can achieve 92.3% and 92.7% on CUB-200-2011 and NABirds, which significantly outperforms the SotA approaches. The source code and pre-trained models are released athttps://github.com/dqshuai/MetaFormer.

64.An End-to-End Approach for Seam Carving Detection using Deep Neural Networks ⬇️

Seam carving is a computational method capable of resizing images for both reduction and expansion based on its content, instead of the image geometry. Although the technique is mostly employed to deal with redundant information, i.e., regions composed of pixels with similar intensity, it can also be used for tampering images by inserting or removing relevant objects. Therefore, detecting such a process is of extreme importance regarding the image security domain. However, recognizing seam-carved images does not represent a straightforward task even for human eyes, and robust computation tools capable of identifying such alterations are very desirable. In this paper, we propose an end-to-end approach to cope with the problem of automatic seam carving detection that can obtain state-of-the-art results. Experiments conducted over public and private datasets with several tampering configurations evidence the suitability of the proposed model.

65.Federated and Generalized Person Re-identification through Domain and Feature Hallucinating ⬇️

In this paper, we study the problem of federated domain generalization (FedDG) for person re-identification (re-ID), which aims to learn a generalized model with multiple decentralized labeled source domains. An empirical method (FedAvg) trains local models individually and averages them to obtain the global model for further local fine-tuning or deploying in unseen target domains. One drawback of FedAvg is neglecting the data distributions of other clients during local training, making the local model overfit local data and producing a poorly-generalized global model. To solve this problem, we propose a novel method, called "Domain and Feature Hallucinating (DFH)", to produce diverse features for learning generalized local and global models. Specifically, after each model aggregation process, we share the Domain-level Feature Statistics (DFS) among different clients without violating data privacy. During local training, the DFS are used to synthesize novel domain statistics with the proposed domain hallucinating, which is achieved by re-weighting DFS with random weights. Then, we propose feature hallucinating to diversify local features by scaling and shifting them to the distribution of the obtained novel domain. The synthesized novel features retain the original pair-wise similarities, enabling us to utilize them to optimize the model in a supervised manner. Extensive experiments verify that the proposed DFH can effectively improve the generalization ability of the global model. Our method achieves the state-of-the-art performance for FedDG on four large-scale re-ID benchmarks.

66.Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detection ⬇️

The recently proposed camouflaged object detection (COD) attempts to segment objects that are visually blended into their surroundings, which is extremely complex and difficult in real-world scenarios. Apart from high intrinsic similarity between the camouflaged objects and their background, the objects are usually diverse in scale, fuzzy in appearance, and even severely occluded. To deal with these problems, we propose a mixed-scale triplet network, \textbf{ZoomNet}, which mimics the behavior of humans when observing vague images, i.e., zooming in and out. Specifically, our ZoomNet employs the zoom strategy to learn the discriminative mixed-scale semantics by the designed scale integration unit and hierarchical mixed-scale unit, which fully explores imperceptible clues between the candidate objects and background surroundings. Moreover, considering the uncertainty and ambiguity derived from indistinguishable textures, we construct a simple yet effective regularization constraint, uncertainty-aware loss, to promote the model to accurately produce predictions with higher confidence in candidate regions. Without bells and whistles, our proposed highly task-friendly model consistently surpasses the existing 23 state-of-the-art methods on four public datasets. Besides, the superior performance over the recent cutting-edge models on the SOD task also verifies the effectiveness and generality of our model. The code will be available at \url{this https URL}.

67.Newton-PnP: Real-time Visual Navigation for Autonomous Toy-Drones ⬇️

The Perspective-n-Point problem aims to estimate the relative pose between a calibrated monocular camera and a known 3D model, by aligning pairs of 2D captured image points to their corresponding 3D points in the model. We suggest an algorithm that runs on weak IoT devices in real-time but still provides provable theoretical guarantees for both running time and correctness. Existing solvers provide only one of these requirements. Our main motivation was to turn the popular DJI's Tello Drone (<90gr, <$100) into an autonomous drone that navigates in an indoor environment with no external human/laptop/sensor, by simply attaching a Raspberry PI Zero (<9gr, <$25) to it. This tiny micro-processor takes as input a real-time video from a tiny RGB camera, and runs our PnP solver on-board. Extensive experimental results, open source code, and a demonstration video are included.

68.Cross Language Image Matching for Weakly Supervised Semantic Segmentation ⬇️

It has been widely known that CAM (Class Activation Map) usually only activates discriminative object regions and falsely includes lots of object-related backgrounds. As only a fixed set of image-level object labels are available to the WSSS (weakly supervised semantic segmentation) model, it could be very difficult to suppress those diverse background regions consisting of open set objects. In this paper, we propose a novel Cross Language Image Matching (CLIMS) framework, based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress closely-related open background regions. In particular, we design object, background region and text label matching losses to guide the model to excite more reasonable object regions for CAM of each category. In addition, we design a co-occurring background suppression loss to prevent the model from activating closely-related background regions, with a predefined set of class-related background text descriptions. These designs enable the proposed CLIMS to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC2012 dataset show that our CLIMS significantly outperforms the previous state-of-the-art methods. Code will be available.

69.Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers ⬇️

Weakly-supervised semantic segmentation (WSSS) with image-level labels is an important and challenging task. Due to the high training efficiency, end-to-end solutions for WSSS have received increasing attention from the community. However, current methods are mainly based on convolutional neural networks and fail to explore the global information properly, thus usually resulting in incomplete object regions. In this paper, to address the aforementioned problem, we introduce Transformers, which naturally integrate global information, to generate more integral initial pseudo labels for end-to-end WSSS. Motivated by the inherent consistency between the self-attention in Transformers and the semantic affinity, we propose an Affinity from Attention (AFA) module to learn semantic affinity from the multi-head self-attention (MHSA) in Transformers. The learned affinity is then leveraged to refine the initial pseudo labels for segmentation. In addition, to efficiently derive reliable affinity labels for supervising AFA and ensure the local consistency of pseudo labels, we devise a Pixel-Adaptive Refinement module that incorporates low-level image appearance information to refine the pseudo labels. We perform extensive experiments and our method achieves 66.0% and 38.9% mIoU on the PASCAL VOC 2012 and MS COCO 2014 datasets, respectively, significantly outperforming recent end-to-end methods and several multi-stage competitors. Code is available at this https URL.

70.A Large-scale Comprehensive Dataset and Copy-overlap Aware Evaluation Protocol for Segment-level Video Copy Detection ⬇️

In this paper, we introduce VCSL (Video Copy Segment Localization), a new comprehensive segment-level annotated video copy dataset. Compared with existing copy detection datasets restricted by either video-level annotation or small-scale, VCSL not only has two orders of magnitude more segment-level labelled data, with 160k realistic video copy pairs containing more than 280k localized copied segment pairs, but also covers a variety of video categories and a wide range of video duration. All the copied segments inside each collected video pair are manually extracted and accompanied by precisely annotated starting and ending timestamps. Alongside the dataset, we also propose a novel evaluation protocol that better measures the prediction accuracy of copy overlapping segments between a video pair and shows improved adaptability in different scenarios. By benchmarking several baseline and state-of-the-art segment-level video copy detection methods with the proposed dataset and evaluation metric, we provide a comprehensive analysis that uncovers the strengths and weaknesses of current approaches, hoping to open up promising directions for future works. The VCSL dataset, metric and benchmark codes are all publicly available at this https URL.

71.Cluster-based Contrastive Disentangling for Generalized Zero-Shot Learning ⬇️

Generalized Zero-Shot Learning (GZSL) aims to recognize both seen and unseen classes by training only the seen classes, in which the instances of unseen classes tend to be biased towards the seen class. In this paper, we propose a Cluster-based Contrastive Disentangling (CCD) method to improve GZSL by alleviating the semantic gap and domain shift problems. Specifically, we first cluster the batch data to form several sets containing similar classes. Then, we disentangle the visual features into semantic-unspecific and semantic-matched variables, and further disentangle the semantic-matched variables into class-shared and class-unique variables according to the clustering results. The disentangled learning module with random swapping and semantic-visual alignment bridges the semantic gap. Moreover, we introduce contrastive learning on semantic-matched and class-unique variables to learn high intra-set and intra-class similarity, as well as inter-set and inter-class discriminability. Then, the generated visual features conform to the underlying characteristics of general images and have strong discriminative information, which alleviates the domain shift problem well. We evaluate our proposed method on four datasets and achieve state-of-the-art results in both conventional and generalized settings.

72.Boosting Crowd Counting via Multifaceted Attention ⬇️

This paper focuses on the challenging crowd counting task. As large-scale variations often exist within crowd images, neither fixed-size convolution kernel of CNN nor fixed-size attention of recent vision transformers can well handle this kind of variation. To address this problem, we propose a Multifaceted Attention Network (MAN) to improve transformer models in local spatial relation encoding. MAN incorporates global attention from a vanilla transformer, learnable local attention, and instance attention into a counting model. Firstly, the local Learnable Region Attention (LRA) is proposed to assign attention exclusively for each feature location dynamically. Secondly, we design the Local Attention Regularization to supervise the training of LRA by minimizing the deviation among the attention for different feature locations. Finally, we provide an Instance Attention mechanism to focus on the most important instances dynamically during training. Extensive experiments on four challenging crowd counting datasets namely ShanghaiTech, UCF-QNRF, JHU++, and NWPU have validated the proposed method. Codes: this https URL.

73.Training privacy-preserving video analytics pipelines by suppressing features that reveal information about private attributes ⬇️

Deep neural networks are increasingly deployed for scene analytics, including to evaluate the attention and reaction of people exposed to out-of-home advertisements. However, the features extracted by a deep neural network that was trained to predict a specific, consensual attribute (e.g. emotion) may also encode and thus reveal information about private, protected attributes (e.g. age or gender). In this work, we focus on such leakage of private information at inference time. We consider an adversary with access to the features extracted by the layers of a deployed neural network and use these features to predict private attributes. To prevent the success of such an attack, we modify the training of the network using a confusion loss that encourages the extraction of features that make it difficult for the adversary to accurately predict private attributes. We validate this training approach on image-based tasks using a publicly available dataset. Results show that, compared to the original network, the proposed PrivateNet can reduce the leakage of private information of a state-of-the-art emotion recognition classifier by 2.88% for gender and by 13.06% for age group, with a minimal effect on task accuracy.

74.Important Object Identification with Semi-Supervised Learning for Autonomous Driving ⬇️

Accurate identification of important objects in the scene is a prerequisite for safe and high-quality decision making and motion planning of intelligent agents (e.g., autonomous vehicles) that navigate in complex and dynamic environments. Most existing approaches attempt to employ attention mechanisms to learn importance weights associated with each object indirectly via various tasks (e.g., trajectory prediction), which do not enforce direct supervision on the importance estimation. In contrast, we tackle this task in an explicit way and formulate it as a binary classification ("important" or "unimportant") problem. We propose a novel approach for important object identification in egocentric driving scenarios with relational reasoning on the objects in the scene. Besides, since human annotations are limited and expensive to obtain, we present a semi-supervised learning pipeline to enable the model to learn from unlimited unlabeled data. Moreover, we propose to leverage the auxiliary tasks of ego vehicle behavior prediction to further improve the accuracy of importance estimation. The proposed approach is evaluated on a public egocentric driving dataset (H3D) collected in complex traffic scenarios. A detailed ablative study is conducted to demonstrate the effectiveness of each model component and the training strategy. Our approach also outperforms rule-based baselines by a large margin.

75.Plant Species Recognition with Optimized 3D Polynomial Neural Networks and Variably Overlapping Time-Coherent Sliding Window ⬇️

Recently, the EAGL-I system was developed to rapidly create massive labeled datasets of plants intended to be commonly used by farmers and researchers to create AI-driven solutions in agriculture. As a result, a publicly available plant species recognition dataset composed of 40,000 images with different sizes consisting of 8 plant species was created with the system in order to demonstrate its capabilities. This paper proposes a novel method, called Variably Overlapping Time-Coherent Sliding Window (VOTCSW), that transforms a dataset composed of images with variable size to a 3D representation with fixed size that is suitable for convolutional neural networks, and demonstrates that this representation is more informative than resizing the images of the dataset to a given size. We theoretically formalized the use cases of the method as well as its inherent properties and we proved that it has an oversampling and a regularization effect on the data. By combining the VOTCSW method with the 3D extension of a recently proposed machine learning model called 1-Dimensional Polynomial Neural Networks, we were able to create a model that achieved a state-of-the-art accuracy of 99.9% on the dataset created by the EAGL-I system, surpassing well-known architectures such as ResNet and Inception. In addition, we created a heuristic algorithm that enables the degree reduction of any pre-trained N-Dimensional Polynomial Neural Network and which compresses it without altering its performance, thus making the model faster and lighter. Furthermore, we established that the currently available dataset could not be used for machine learning in its present form, due to a substantial class imbalance between the training set and the test set. Hence, we created a specific preprocessing and a model development framework that enabled us to improve the accuracy from 49.23% to 99.9%.

76.A Quality Index Metric and Method for Online Self-Assessment of Autonomous Vehicles Sensory Perception ⬇️

Perception is critical to autonomous driving safety. Camera-based object detection is one of the most important methods for autonomous vehicle perception. Current camera-based object detection solutions for autonomous driving cannot provide feedback on the detection performance for each frame. We propose an evaluation metric, namely the perception quality index (PQI), to assess the camera-based object detection algorithm performance and provide the perception quality feedback frame by frame. The method of the PQI generation is by combining the fine-grained saliency map intensity with the object detection algorithm's output results. Furthermore, we developed a superpixel-based attention network (SPA-NET) to predict the proposed PQI evaluation metric by using raw image pixels and superpixels as input. The proposed evaluation metric and prediction network are tested on three open-source datasets. The proposed evaluation metric can correctly assess the camera-based perception quality under the autonomous driving environment according to the experiment results. The network regression R-square values determine the comparison among models. It is shown that a Perception Quality Index is useful in self-evaluating a cameras visual scene perception.

77.Online Learning of Reusable Abstract Models for Object Goal Navigation ⬇️

In this paper, we present a novel approach to incrementally learn an Abstract Model of an unknown environment, and show how an agent can reuse the learned model for tackling the Object Goal Navigation task. The Abstract Model is a finite state machine in which each state is an abstraction of a state of the environment, as perceived by the agent in a certain position and orientation. The perceptions are high-dimensional sensory data (e.g., RGB-D images), and the abstraction is reached by exploiting image segmentation and the Taskonomy model bank. The learning of the Abstract Model is accomplished by executing actions, observing the reached state, and updating the Abstract Model with the acquired information. The learned models are memorized by the agent, and they are reused whenever it recognizes to be in an environment that corresponds to the stored model. We investigate the effectiveness of the proposed approach for the Object Goal Navigation task, relying on public benchmarks. Our results show that the reuse of learned Abstract Models can boost performance on Object Goal Navigation.

78.Style-ERD: Responsive and Coherent Online Motion Style Transfer ⬇️

Motion style transfer is a common method for enriching character animation. Motion style transfer algorithms are often designed for offline settings where motions are processed in segments. However, for online animation applications, such as realtime avatar animation from motion capture, motions need to be processed as a stream with minimal latency. In this work, we realize a flexible, high-quality motion style transfer method for this setting. We propose a novel style transfer model, Style-ERD, to stylize motions in an online manner with an Encoder-Recurrent-Decoder structure, along with a novel discriminator that combines feature attention and temporal attention. Our method stylizes motions into multiple target styles with a unified model. Although our method targets online settings, it outperforms previous offline methods in motion realism and style expressiveness and provides significant gains in runtime efficiency

79.Show Me What and Tell Me How: Video Synthesis via Multimodal Conditioning ⬇️

Most methods for conditional video synthesis use a single modality as the condition. This comes with major limitations. For example, it is problematic for a model conditioned on an image to generate a specific motion trajectory desired by the user since there is no means to provide motion information. Conversely, language information can describe the desired motion, while not precisely defining the content of the video. This work presents a multimodal video generation framework that benefits from text and images provided jointly or separately. We leverage the recent progress in quantized representations for videos and apply a bidirectional transformer with multiple modalities as inputs to predict a discrete video representation. To improve video quality and consistency, we propose a new video token trained with self-learning and an improved mask-prediction algorithm for sampling video tokens. We introduce text augmentation to improve the robustness of the textual representation and diversity of generated videos. Our framework can incorporate various visual modalities, such as segmentation masks, drawings, and partially occluded images. It can generate much longer sequences than the one used for training. In addition, our model can extract visual information as suggested by the text prompt, e.g., "an object in image one is moving northeast", and generate corresponding videos. We run evaluations on three public datasets and a newly collected dataset labeled with facial attributes, achieving state-of-the-art generation results on all four.

80.UVCGAN: UNet Vision Transformer cycle-consistent GAN for unpaired image-to-image translation ⬇️

Image-to-image translation has broad applications in art, design, and scientific simulations. The original CycleGAN model emphasizes one-to-one mapping via a cycle-consistent loss, while more recent works promote one-to-many mapping to boost the diversity of the translated images. With scientific simulation and one-to-one needs in mind, this work examines if equipping CycleGAN with a vision transformer (ViT) and employing advanced generative adversarial network (GAN) training techniques can achieve better performance. The resulting UNet ViT Cycle-consistent GAN (UVCGAN) model is compared with previous best-performing models on open benchmark image-to-image translation datasets, Selfie2Anime and CelebA. UVCGAN performs better and retains a strong correlation between the original and translated images. An accompanying ablation study shows that the gradient penalty and BERT-like pre-training also contribute to the improvement.~To promote reproducibility and open science, the source code, hyperparameter configurations, and pre-trained model will be made available at: this https URL.

81.Building 3D Generative Models from Minimal Data ⬇️

We propose a method for constructing generative models of 3D objects from a single 3D mesh and improving them through unsupervised low-shot learning from 2D images. Our method produces a 3D morphable model that represents shape and albedo in terms of Gaussian processes. Whereas previous approaches have typically built 3D morphable models from multiple high-quality 3D scans through principal component analysis, we build 3D morphable models from a single scan or template. As we demonstrate in the face domain, these models can be used to infer 3D reconstructions from 2D data (inverse graphics) or 3D data (registration). Specifically, we show that our approach can be used to perform face recognition using only a single 3D template (one scan total, not one per person). We extend our model to a preliminary unsupervised learning framework that enables the learning of the distribution of 3D faces using one 3D template and a small number of 2D images. This approach could also provide a model for the origins of face perception in human infants, who appear to start with an innate face template and subsequently develop a flexible system for perceiving the 3D structure of any novel face from experience with only 2D images of a relatively small number of familiar faces.

82.Structured Pruning is All You Need for Pruning CNNs at Initialization ⬇️

Pruning is a popular technique for reducing the model size and computational cost of convolutional neural networks (CNNs). However, a slow retraining or fine-tuning procedure is often required to recover the accuracy loss caused by pruning. Recently, a new research direction on weight pruning, pruning-at-initialization (PAI), is proposed to directly prune CNNs before training so that fine-tuning or retraining can be avoided. While PAI has shown promising results in reducing the model size, existing approaches rely on fine-grained weight pruning which requires unstructured sparse matrix computation, making it difficult to achieve real speedup in practice unless the sparsity is very high.
This work is the first to show that fine-grained weight pruning is in fact not necessary for PAI. Instead, the layerwise compression ratio is the main critical factor to determine the accuracy of a CNN model pruned at initialization. Based on this key observation, we propose PreCropping, a structured hardware-efficient model compression scheme. PreCropping directly compresses the model at the channel level following the layerwise compression ratio. Compared to weight pruning, the proposed scheme is regular and dense in both storage and computation without sacrificing accuracy. In addition, since PreCropping compresses CNNs at initialization, the computational and memory costs of CNNs are reduced for both training and inference on commodity hardware. We empirically demonstrate our approaches on several modern CNN architectures, including ResNet, ShuffleNet, and MobileNet for both CIFAR-10 and ImageNet.

83.Cartoon-texture evolution for two-region image segmentation ⬇️

Two-region image segmentation is the process of dividing an image into two regions of interest, i.e., the foreground and the background. To this aim, Chan et al. [Chan, Esedoglu, Nikolova, SIAM Journal on Applied Mathematics 66(5), 1632-1648, 2006] designed a model well suited for smooth images. One drawback of this model is that it may produce a bad segmentation when the image contains oscillatory components. Based on a cartoon-texture decomposition of the image to be segmented, we propose a new model that is able to produce an accurate segmentation of images also containing noise or oscillatory information like texture. The novel model leads to a non-smooth constrained optimization problem which we solve by means of the ADMM method. The convergence of the numerical scheme is also proved. Several experiments on smooth, noisy, and textural images show the effectiveness of the proposed model.

84.Graph Neural Networks for Image Classification and Reinforcement Learning using Graph representations ⬇️

In this paper, we will evaluate the performance of graph neural networks in two distinct domains: computer vision and reinforcement learning. In the computer vision section, we seek to learn whether a novel non-redundant representation for images as graphs can improve performance over trivial pixel to node mapping on a graph-level prediction graph, specifically image classification. For the reinforcement learning section, we seek to learn if explicitly modeling solving a Rubik's cube as a graph problem can improve performance over a standard model-free technique with no inductive bias.

85.CoNIC Solution ⬇️

Nuclei segmentation and classification has been a challenge due to the high inter-class similarity and intra-class variability. Thus, a large-scale annotation and a specially-designed algorithm are needed to solve this problem. Lizard is therefore a great promotion in this area, containing around half a million nuclei annotated. In this paper, we propose a two-stage pipeline used in the CoNIC competition, which achieves much better results than the baseline method. We adopt a similar model as the original baseline method: HoVerNet, as the segmentaion model and then develop a new classification model to fine-tune the classification results. Code for this method will be made public soon. This is a conic solution in testing.

86.FloorGenT: Generative Vector Graphic Model of Floor Plans for Robotics ⬇️

Floor plans are the basis of reasoning in and communicating about indoor environments. In this paper, we show that by modelling floor plans as sequences of line segments seen from a particular point of view, recent advances in autoregressive sequence modelling can be leveraged to model and predict floor plans. The line segments are canonicalized and translated to sequence of tokens and an attention-based neural network is used to fit a one-step distribution over next tokens. We fit the network to sequences derived from a set of large-scale floor plans, and demonstrate the capabilities of the model in four scenarios: novel floor plan generation, completion of partially observed floor plans, generation of floor plans from simulated sensor data, and finally, the applicability of a floor plan model in predicting the shortest distance with partial knowledge of the environment.

87.Joint brain tumor segmentation from multi MR sequences through a deep convolutional neural network ⬇️

Brain tumor segmentation is highly contributive in diagnosing and treatment planning. The manual brain tumor delineation is a time-consuming and tedious task and varies depending on the radiologists skill. Automated brain tumor segmentation is of high importance, and does not depend on either inter or intra-observation. The objective of this study is to automate the delineation of brain tumors from the FLAIR, T1 weighted, T2 weighted, and T1 weighted contrast-enhanced MR sequences through a deep learning approach, with a focus on determining which MR sequence alone or which combination thereof would lead to the highest accuracy therein.

88.A novel shape-based loss function for machine learning-based seminal organ segmentation in medical imaging ⬇️

Automated medical image segmentation is an essential task to aid/speed up diagnosis and treatment procedures in clinical practices. Deep convolutional neural networks have exhibited promising performance in accurate and automatic seminal segmentation. For segmentation tasks, these methods normally rely on minimizing a cost/loss function that is designed to maximize the overlap between the estimated target and the ground-truth mask delineated by the experts. A simple loss function based on the degrees of overlap (i.e., Dice metric) would not take into account the underlying shape and morphology of the target subject, as well as its realistic/natural variations; therefore, suboptimal segmentation results would be observed in the form of islands of voxels, holes, and unrealistic shapes or deformations. In this light, many studies have been conducted to refine/post-process the segmentation outcome and consider an initial guess as prior knowledge to avoid outliers and/or unrealistic estimations. In this study, a novel shape-based cost function is proposed which encourages/constrains the network to learn/capture the underlying shape features in order to generate a valid/realistic estimation of the target structure. To this end, the Principal Component Analysis (PCA) was performed on a vectorized training dataset to extract eigenvalues and eigenvectors of the target subjects. The key idea was to use the reconstruction weights to discriminate valid outcomes from outliers/erroneous estimations.

89.Predicting Bearings' Degradation Stages for Predictive Maintenance in the Pharmaceutical Industry ⬇️

In the pharmaceutical industry, the maintenance of production machines must be audited by the regulator. In this context, the problem of predictive maintenance is not when to maintain a machine, but what parts to maintain at a given point in time. The focus shifts from the entire machine to its component parts and prediction becomes a classification problem. In this paper, we focus on rolling-elements bearings and we propose a framework for predicting their degradation stages automatically. Our main contribution is a k-means bearing lifetime segmentation method based on high-frequency bearing vibration signal embedded in a latent low-dimensional subspace using an AutoEncoder. Given high-frequency vibration data, our framework generates a labeled dataset that is used to train a supervised model for bearing degradation stage detection. Our experimental results, based on the FEMTO Bearing dataset, show that our framework is scalable and that it provides reliable and actionable predictions for a range of different bearings.

90.Augmented Reality and Robotics: A Survey and Taxonomy for AR-enhanced Human-Robot Interaction and Robotic Interfaces ⬇️

This paper contributes to a taxonomy of augmented reality and robotics based on a survey of 460 research papers. Augmented and mixed reality (AR/MR) have emerged as a new way to enhance human-robot interaction (HRI) and robotic interfaces (e.g., actuated and shape-changing interfaces). Recently, an increasing number of studies in HCI, HRI, and robotics have demonstrated how AR enables better interactions between people and robots. However, often research remains focused on individual explorations and key design strategies, and research questions are rarely analyzed systematically. In this paper, we synthesize and categorize this research field in the following dimensions: 1) approaches to augmenting reality; 2) characteristics of robots; 3) purposes and benefits; 4) classification of presented information; 5) design components and strategies for visual augmentation; 6) interaction techniques and modalities; 7) application domains; and 8) evaluation strategies. We formulate key challenges and opportunities to guide and inform future research in AR and robotics.

91.Maximizing Conditional Independence for Unsupervised Domain Adaptation ⬇️

Unsupervised domain adaptation studies how to transfer a learner from a labeled source domain to an unlabeled target domain with different distributions. Existing methods mainly focus on matching the marginal distributions of the source and target domains, which probably lead a misalignment of samples from the same class but different domains. In this paper, we deal with this misalignment by achieving the class-conditioned transferring from a new perspective. We aim to maximize the conditional independence of feature and domain given class in the reproducing kernel Hilbert space. The optimization of the conditional independence measure can be viewed as minimizing a surrogate of a certain mutual information between feature and domain. An interpretable empirical estimation of the conditional dependence is deduced and connected with the unconditional case. Besides, we provide an upper bound on the target error by taking the class-conditional distribution into account, which provides a new theoretical insight for most class-conditioned transferring methods. In addition to unsupervised domain adaptation, we extend our method to the multi-source scenario in a natural and elegant way. Extensive experiments on four benchmarks validate the effectiveness of the proposed models in both unsupervised domain adaptation and multiple source domain adaptation.

92.Undersampled MRI Reconstruction with Side Information-Guided Normalisation ⬇️

Magnetic resonance (MR) images exhibit various contrasts and appearances based on factors such as different acquisition protocols, views, manufacturers, scanning parameters, etc. This generally accessible appearance-related side information affects deep learning-based undersampled magnetic resonance imaging (MRI) reconstruction frameworks, but has been overlooked in the majority of current works. In this paper, we investigate the use of such side information as normalisation parameters in a convolutional neural network (CNN) to improve undersampled MRI reconstruction. Specifically, a Side Information-Guided Normalisation (SIGN) module, containing only few layers, is proposed to efficiently encode the side information and output the normalisation parameters. We examine the effectiveness of such a module on two popular reconstruction architectures, D5C5 and OUCR. The experimental results on both brain and knee images under various acceleration rates demonstrate that the proposed method improves on its corresponding baseline architectures with a significant margin.

93.Differentially Private Federated Learning with Local Regularization and Sparsification ⬇️

User-level differential privacy (DP) provides certifiable privacy guarantees to the information that is specific to any user's data in federated learning. Existing methods that ensure user-level DP come at the cost of severe accuracy decrease. In this paper, we study the cause of model performance degradation in federated learning under user-level DP guarantee. We find the key to solving this issue is to naturally restrict the norm of local updates before executing operations that guarantee DP. To this end, we propose two techniques, Bounded Local Update Regularization and Local Update Sparsification, to increase model quality without sacrificing privacy. We provide theoretical analysis on the convergence of our framework and give rigorous privacy guarantees. Extensive experiments show that our framework significantly improves the privacy-utility trade-off over the state-of-the-arts for federated learning with user-level DP guarantee.

94.Virtual vs. Reality: External Validation of COVID-19 Classifiers using XCAT Phantoms for Chest Computed Tomography ⬇️

Research studies of artificial intelligence models in medical imaging have been hampered by poor generalization. This problem has been especially concerning over the last year with numerous applications of deep learning for COVID-19 diagnosis. Virtual imaging trials (VITs) could provide a solution for objective evaluation of these models. In this work utilizing the VITs, we created the CVIT-COVID dataset including 180 virtually imaged computed tomography (CT) images from simulated COVID-19 and normal phantom models under different COVID-19 morphology and imaging properties. We evaluated the performance of an open-source, deep-learning model from the University of Waterloo trained with multi-institutional data and an in-house model trained with the open clinical dataset called MosMed. We further validated the model's performance against open clinical data of 305 CT images to understand virtual vs. real clinical data performance. The open-source model was published with nearly perfect performance on the original Waterloo dataset but showed a consistent performance drop in external testing on another clinical dataset (AUC=0.77) and our simulated CVIT-COVID dataset (AUC=0.55). The in-house model achieved an AUC of 0.87 while testing on the internal test set (MosMed test set). However, performance dropped to an AUC of 0.65 and 0.69 when evaluated on clinical and our simulated CVIT-COVID dataset. The VIT framework offered control over imaging conditions, allowing us to show there was no change in performance as CT exposure was changed from 28.5 to 57 mAs. The VIT framework also provided voxel-level ground truth, revealing that performance of in-house model was much higher at AUC=0.87 for diffuse COVID-19 infection size >2.65% lung volume versus AUC=0.52 for focal disease with <2.65% volume. The virtual imaging framework enabled these uniquely rigorous analyses of model performance.

95.A Perspective on Robotic Telepresence and Teleoperation using Cognition: Are we there yet? ⬇️

Telepresence and teleoperation robotics have attracted a great amount of attention in the last 10 years. With the Artificial Intelligence (AI) revolution already being started, we can see a wide range of robotic applications being realized. Intelligent robotic systems are being deployed both in industrial and domestic environments. Telepresence is the idea of being present in a remote location virtually or via robotic avatars. Similarly, the idea of operating a robot from a remote location for various tasks is called teleoperation. These technologies find significant application in health care, education, surveillance, disaster recovery, and corporate/government sectors. But question still remains about their maturity, security and safety levels. We also need to think about enhancing the user experience and trust in such technologies going into the next generation of computing.

96.Evaluation of Interpretability Methods and Perturbation Artifacts in Deep Neural Networks ⬇️

The challenge of interpreting predictions from deep neural networks has prompted the development of numerous interpretability methods. Many of interpretability methods attempt to quantify the importance of input features with respect to the class probabilities, and are called importance estimators or saliency maps. A popular approach to evaluate such interpretability methods is to perturb input features deemed important for predictions and observe the decrease in accuracy. However, perturbation-based evaluation methods may confound the sources of accuracy degradation. We conduct computational experiments that allow to empirically estimate the $\textit{fidelity}$ of interpretability methods and the contribution of perturbation artifacts. All considered importance estimators clearly outperform a random baseline, which contradicts the findings of ROAR [arXiv:1806.10758]. We further compare our results to the crop-and-resize evaluation framework [arXiv:1705.07857], which are largely in agreement. Our study suggests that we can estimate the impact of artifacts and thus empirically evaluate interpretability methods without retraining.

97.Machine Learning Applications in Diagnosis, Treatment and Prognosis of Lung Cancer ⬇️

The recent development of imaging and sequencing technologies enables systematic advances in the clinical study of lung cancer. Meanwhile, the human mind is limited in effectively handling and fully utilizing the accumulation of such enormous amounts of data. Machine learning-based approaches play a critical role in integrating and analyzing these large and complex datasets, which have extensively characterized lung cancer through the use of different perspectives from these accrued data. In this article, we provide an overview of machine learning-based approaches that strengthen the varying aspects of lung cancer diagnosis and therapy, including early detection, auxiliary diagnosis, prognosis prediction and immunotherapy practice. Moreover, we highlight the challenges and opportunities for future applications of machine learning in lung cancer.

98.Rib Suppression in Digital Chest Tomosynthesis ⬇️

Digital chest tomosynthesis (DCT) is a technique to produce sectional 3D images of a human chest for pulmonary disease screening, with 2D X-ray projections taken within an extremely limited range of angles. However, under the limited angle scenario, DCT contains strong artifacts caused by the presence of ribs, jamming the imaging quality of the lung area. Recently, great progress has been achieved for rib suppression in a single X-ray image, to reveal a clearer lung texture. We firstly extend the rib suppression problem to the 3D case at the software level. We propose a $\textbf{T}$omosynthesis $\textbf{RI}$b Su$\textbf{P}$pression and $\textbf{L}$ung $\textbf{E}$nhancement $\textbf{Net}$work (TRIPLE-Net) to model the 3D rib component and provide a rib-free DCT. TRIPLE-Net takes the advantages from both 2D and 3D domains, which model the ribs in DCT with the exact FBP procedure and 3D depth information, respectively. The experiments on simulated datasets and clinical data have shown the effectiveness of TRIPLE-Net to preserve lung details as well as improve the imaging quality of pulmonary diseases. Finally, an expert user study confirms our findings.

99.DrawingInStyles: Portrait Image Generation and Editing with Spatially Conditioned StyleGAN ⬇️

The research topic of sketch-to-portrait generation has witnessed a boost of progress with deep learning techniques. The recently proposed StyleGAN architectures achieve state-of-the-art generation ability but the original StyleGAN is not friendly for sketch-based creation due to its unconditional generation nature. To address this issue, we propose a direct conditioning strategy to better preserve the spatial information under the StyleGAN framework. Specifically, we introduce Spatially Conditioned StyleGAN (SC-StyleGAN for short), which explicitly injects spatial constraints to the original StyleGAN generation process. We explore two input modalities, sketches and semantic maps, which together allow users to express desired generation results more precisely and easily. Based on SC-StyleGAN, we present DrawingInStyles, a novel drawing interface for non-professional users to easily produce high-quality, photo-realistic face images with precise control, either from scratch or editing existing ones. Qualitative and quantitative evaluations show the superior generation ability of our method to existing and alternative solutions. The usability and expressiveness of our system are confirmed by a user study.

100.MaxDropoutV2: An Improved Method to Drop out Neurons in Convolutional Neural Networks ⬇️

In the last decade, exponential data growth supplied the machine learning-based algorithms' capacity and enabled their usage in daily life activities. Additionally, such an improvement is partially explained due to the advent of deep learning techniques, i.e., stacks of simple architectures that end up in more complex models. Although both factors produce outstanding results, they also pose drawbacks regarding the learning process since training complex models denotes an expensive task and results are prone to overfit the training data. A supervised regularization technique called MaxDropout was recently proposed to tackle the latter, providing several improvements concerning traditional regularization approaches. In this paper, we present its improved version called MaxDropoutV2. Results considering two public datasets show that the model performs faster than the standard version and, in most cases, provides more accurate results.

101.A Novel Dual Dense Connection Network for Video Super-resolution ⬇️

Video super-resolution (VSR) refers to the reconstruction of high-resolution (HR) video from the corresponding low-resolution (LR) video. Recently, VSR has received increasing attention. In this paper, we propose a novel dual dense connection network that can generate high-quality super-resolution (SR) results. The input frames are creatively divided into reference frame, pre-temporal group and post-temporal group, representing information in different time periods. This grouping method provides accurate information of different time periods without causing time information disorder. Meanwhile, we produce a new loss function, which is beneficial to enhance the convergence ability of the model. Experiments show that our model is superior to other advanced models in Vid4 datasets and SPMCS-11 datasets.

102.Towards Efficient and Scalable Sharpness-Aware Minimization ⬇️

Recently, Sharpness-Aware Minimization (SAM), which connects the geometry of the loss landscape and generalization, has demonstrated significant performance boosts on training large-scale models such as vision transformers. However, the update rule of SAM requires two sequential (non-parallelizable) gradient computations at each step, which can double the computational overhead. In this paper, we propose a novel algorithm LookSAM - that only periodically calculates the inner gradient ascent, to significantly reduce the additional training cost of SAM. The empirical results illustrate that LookSAM achieves similar accuracy gains to SAM while being tremendously faster - it enjoys comparable computational complexity with first-order optimizers such as SGD or Adam. To further evaluate the performance and scalability of LookSAM, we incorporate a layer-wise modification and perform experiments in the large-batch training scenario, which is more prone to converge to sharp local minima. We are the first to successfully scale up the batch size when training Vision Transformers (ViTs). With a 64k batch size, we are able to train ViTs from scratch in minutes while maintaining competitive performance.

103.High-resolution Coastline Extraction in SAR Images via MISP-GGD Superpixel Segmentation ⬇️

High accuracy coastline/shoreline extraction from SAR imagery is a crucial step in a number of maritime and coastal monitoring applications. We present a method based on image segmentation using the Generalised Gamma Mixture Model superpixel algorithm (MISP-GGD). MISP-GGD produces superpixels adhering with great accuracy to object edges in the image, such as the coastline. Unsupervised clustering of the generated superpixels according to textural and radiometric features allows for generation of a land/water mask from which a highly accurate coastline can be extracted. We present results of our proposed method on a number of SAR images of varying characteristics.

104.IDmUNet: A new image decomposition induced network for sparse feature segmentation ⬇️

UNet and its variants are among the most popular methods for medical image segmentation. Despite their successes in task generality, most of them consider little mathematical modeling behind specific applications. In this paper, we focus on the sparse feature segmentation task and make a task-oriented network design, in which the target objects are sparsely distributed and the background is hard to be mathematically modeled. We start from an image decomposition model with sparsity regularization, and propose a deep unfolding network, namely IDNet, based on an iterative solver, scaled alternating direction method of multipliers (scaled-ADMM). The IDNet splits raw inputs into double feature layers. Then a new task-oriented segmentation network is constructed, dubbed as IDmUNet, based on the proposed IDNets and a mini-UNet. Because of the sparsity prior and deep unfolding method in the structure design, this IDmUNet combines the advantages of mathematical modeling and data-driven approaches. Firstly, our approach has mathematical interpretability and can achieve favorable performance with far fewer learnable parameters. Secondly, our IDmUNet is robust in a simple end-to-end training with explainable behaviors. In the experiments of retinal vessel segmentation (RVS), IDmUNet produces the state-of-the-art results with only 0.07m parameters, whereas SA-UNet, one of the latest variants of UNet, contains 0.54m and the original UNet 31.04m. Moreover, the training procedure of our network converges faster without overfitting phenomenon. This decomposition-based network construction strategy can be generalized to other problems with mathematically clear targets and complicated unclear backgrounds.

105.Audio-visual speech separation based on joint feature representation with cross-modal attention ⬇️

Multi-modal based speech separation has exhibited a specific advantage on isolating the target character in multi-talker noisy environments. Unfortunately, most of current separation strategies prefer a straightforward fusion based on feature learning of each single modality, which is far from sufficient consideration of inter-relationships between modalites. Inspired by learning joint feature representations from audio and visual streams with attention mechanism, in this study, a novel cross-modal fusion strategy is proposed to benefit the whole framework with semantic correlations between different modalities. To further improve audio-visual speech separation, the dense optical flow of lip motion is incorporated to strengthen the robustness of visual representation. The evaluation of the proposed work is performed on two public audio-visual speech separation benchmark datasets. The overall improvement of the performance has demonstrated that the additional motion network effectively enhances the visual representation of the combined lip images and audio signal, as well as outperforming the baseline in terms of all metrics with the proposed cross-modal fusion.

106.Ensemble Knowledge Guided Sub-network Search and Fine-tuning for Filter Pruning ⬇️

Conventional NAS-based pruning algorithms aim to find the sub-network with the best validation performance. However, validation performance does not successfully represent test performance, i.e., potential performance. Also, although fine-tuning the pruned network to restore the performance drop is an inevitable process, few studies have handled this issue. This paper proposes a novel sub-network search and fine-tuning method that is named Ensemble Knowledge Guidance (EKG). First, we experimentally prove that the fluctuation of the loss landscape is an effective metric to evaluate the potential performance. In order to search a sub-network with the smoothest loss landscape at a low cost, we propose a pseudo-supernet built by an ensemble sub-network knowledge distillation. Next, we propose a novel fine-tuning that re-uses the information of the search phase. We store the interim sub-networks, that is, the by-products of the search phase, and transfer their knowledge into the pruned network. Note that EKG is easy to be plugged-in and computationally efficient. For example, in the case of ResNet-50, about 45% of FLOPS is removed without any performance drop in only 315 GPU hours. The implemented code is available at this https URL.

107.How to Train Unstable Looped Tensor Network ⬇️

A rising problem in the compression of Deep Neural Networks is how to reduce the number of parameters in convolutional kernels and the complexity of these layers by low-rank tensor approximation. Canonical polyadic tensor decomposition (CPD) and Tucker tensor decomposition (TKD) are two solutions to this problem and provide promising results. However, CPD often fails due to degeneracy, making the networks unstable and hard to fine-tune. TKD does not provide much compression if the core tensor is big. This motivates using a hybrid model of CPD and TKD, a decomposition with multiple Tucker models with small core tensor, known as block term decomposition (BTD). This paper proposes a more compact model that further compresses the BTD by enforcing core tensors in BTD identical. We establish a link between the BTD with shared parameters and a looped chain tensor network (TC). Unfortunately, such strongly constrained tensor networks (with loop) encounter severe numerical instability, as proved by y (Landsberg, 2012) and (Handschuh, 2015a). We study perturbation of chain tensor networks, provide interpretation of instability in TC, demonstrate the problem. We propose novel methods to gain the stability of the decomposition results, keep the network robust and attain better approximation. Experimental results will confirm the superiority of the proposed methods in compression of well-known CNNs, and TC decomposition under challenging scenarios

108.Geodesic Gramian Denoising Applied to the Images Contaminated With Noise Sampled From Diverse Probability Distributions ⬇️

As quotidian use of sophisticated cameras surges, people in modern society are more interested in capturing fine-quality images. However, the quality of the images might be inferior to people's expectations due to the noise contamination in the images. Thus, filtering out the noise while preserving vital image features is an essential requirement. Current existing denoising methods have their own assumptions on the probability distribution in which the contaminated noise is sampled for the method to attain its expected denoising performance. In this paper, we utilize our recent Gramian-based filtering scheme to remove noise sampled from five prominent probability distributions from selected images. This method preserves image smoothness by adopting patches partitioned from the image, rather than pixels, and retains vital image features by performing denoising on the manifold underlying the patch space rather than in the image domain. We validate its denoising performance, using three benchmark computer vision test images applied to two state-of-the-art denoising methods, namely BM3D and K-SVD.

109.Concept-based Explanations for Out-Of-Distribution Detectors ⬇️

Out-of-distribution (OOD) detection plays a crucial role in ensuring the safe deployment of deep neural network (DNN) classifiers. While a myriad of methods have focused on improving the performance of OOD detectors, a critical gap remains in interpreting their decisions. We help bridge this gap by providing explanations for OOD detectors based on learned high-level concepts. We first propose two new metrics for assessing the effectiveness of a particular set of concepts for explaining OOD detectors: 1) detection completeness, which quantifies the sufficiency of concepts for explaining an OOD-detector's decisions, and 2) concept separability, which captures the distributional separation between in-distribution and OOD data in the concept space. Based on these metrics, we propose a framework for learning a set of concepts that satisfy the desired properties of detection completeness and concept separability and demonstrate the framework's effectiveness in providing concept-based explanations for diverse OOD techniques. We also show how to identify prominent concepts that contribute to the detection results via a modified Shapley value-based importance score.

110.Improving the Energy Efficiency and Robustness of tinyML Computer Vision using Log-Gradient Input Images ⬇️

This paper studies the merits of applying log-gradient input images to convolutional neural networks (CNNs) for tinyML computer vision (CV). We show that log gradients enable: (i) aggressive 1.5-bit quantization of first-layer inputs, (ii) potential CNN resource reductions, and (iii) inherent robustness to illumination changes (1.7% accuracy loss across 1/32...8 brightness variation vs. up to 10% for JPEG). We establish these results using the PASCAL RAW image data set and through a combination of experiments using neural architecture search and a fixed three-layer network. The latter reveal that training on log-gradient images leads to higher filter similarity, making the CNN more prunable. The combined benefits of aggressive first-layer quantization, CNN resource reductions, and operation without tight exposure control and image signal processing (ISP) are helpful for pushing tinyML CV toward its ultimate efficiency limits.

111.BoostMIS: Boosting Medical Image Semi-supervised Learning with Adaptive Pseudo Labeling and Informative Active Annotation ⬇️

In this paper, we propose a novel semi-supervised learning (SSL) framework named BoostMIS that combines adaptive pseudo labeling and informative active annotation to unleash the potential of medical image SSL models: (1) BoostMIS can adaptively leverage the cluster assumption and consistency regularization of the unlabeled data according to the current learning status. This strategy can adaptively generate one-hot ``hard'' labels converted from task model predictions for better task model training. (2) For the unselected unlabeled images with low confidence, we introduce an Active learning (AL) algorithm to find the informative samples as the annotation candidates by exploiting virtual adversarial perturbation and model's density-aware entropy. These informative candidates are subsequently fed into the next training cycle for better SSL label propagation. Notably, the adaptive pseudo-labeling and informative active annotation form a learning closed-loop that are mutually collaborative to boost medical image SSL. To verify the effectiveness of the proposed method, we collected a metastatic epidural spinal cord compression (MESCC) dataset that aims to optimize MESCC diagnosis and classification for improved specialist referral and treatment. We conducted an extensive experimental study of BoostMIS on MESCC and another public dataset COVIDx. The experimental results verify our framework's effectiveness and generalisability for different medical image datasets with a significant improvement over various state-of-the-art methods.

112.Cellular Segmentation and Composition in Routine Histology Images using Deep Learning ⬇️

Identification and quantification of nuclei in colorectal cancer haematoxylin &amp; eosin (H&amp;E) stained histology images is crucial to prognosis and patient management. In computational pathology these tasks are referred to as nuclear segmentation, classification and composition and are used to extract meaningful interpretable cytological and architectural features for downstream analysis. The CoNIC challenge poses the task of automated nuclei segmentation, classification and composition into six different types of nuclei from the largest publicly known nuclei dataset - Lizard. In this regard, we have developed pipelines for the prediction of nuclei segmentation using HoVer-Net and ALBRT for cellular composition. On testing on the preliminary test set, HoVer-Net achieved a PQ of 0.58, a PQ+ of 0.58 and finally a mPQ+ of 0.35. For the prediction of cellular composition with ALBRT on the preliminary test set, we achieved an overall $R^2$ score of 0.53, consisting of 0.84 for lymphocytes, 0.70 for epithelial cells, 0.70 for plasma and .060 for eosinophils.

113.ARM 4-BIT PQ: SIMD-based Acceleration for Approximate Nearest Neighbor Search on ARM ⬇️

We accelerate the 4-bit product quantization (PQ) on the ARM architecture. Notably, the drastic performance of the conventional 4-bit PQ strongly relies on x64-specific SIMD register, such as AVX2; hence, we cannot yet achieve such good performance on ARM. To fill this gap, we first bundle two 128-bit registers as one 256-bit component. We then apply shuffle operations for each using the ARM-specific NEON instruction. By making this simple but critical modification, we achieve a dramatic speedup for the 4-bit PQ on an ARM architecture. Experiments show that the proposed method consistently achieves a 10x improvement over the naive PQ with the same accuracy.