20220307.md

ArXiv cs.CV --Mon, 7 Mar 2022

1.HyperTransformer: A Textural and Spectral Feature Fusion Transformer for Pansharpening ⬇️

Pansharpening aims to fuse a registered high-resolution panchromatic image (PAN) with a low-resolution hyperspectral image (LR-HSI) to generate an enhanced HSI with high spectral and spatial resolution. Existing pansharpening approaches neglect using an attention mechanism to transfer HR texture features from PAN to LR-HSI features, resulting in spatial and spectral distortions. In this paper, we present a novel attention mechanism for pansharpening called HyperTransformer, in which features of LR-HSI and PAN are formulated as queries and keys in a transformer, respectively. HyperTransformer consists of three main modules, namely two separate feature extractors for PAN and HSI, a multi-head feature soft attention module, and a spatial-spectral feature fusion module. Such a network improves both spatial and spectral quality measures of the pansharpened HSI by learning cross-feature space dependencies and long-range details of PAN and LR-HSI. Furthermore, HyperTransformer can be utilized across multiple spatial scales at the backbone for obtaining improved performance. Extensive experiments conducted on three widely used datasets demonstrate that HyperTransformer achieves significant improvement over the state-of-the-art methods on both spatial and spectral quality measures. Implementation code and pre-trained weights can be accessed at this https URL.

2.Pedestrian Stop and Go Forecasting with Hybrid Feature Fusion ⬇️

Forecasting pedestrians' future motions is essential for autonomous driving systems to safely navigate in urban areas. However, existing prediction algorithms often overly rely on past observed trajectories and tend to fail around abrupt dynamic changes, such as when pedestrians suddenly start or stop walking. We suggest that predicting these highly non-linear transitions should form a core component to improve the robustness of motion prediction algorithms. In this paper, we introduce the new task of pedestrian stop and go forecasting. Considering the lack of suitable existing datasets for it, we release TRANS, a benchmark for explicitly studying the stop and go behaviors of pedestrians in urban traffic. We build it from several existing datasets annotated with pedestrians' walking motions, in order to have various scenarios and behaviors. We also propose a novel hybrid model that leverages pedestrian-specific and scene features from several modalities, both video sequences and high-level attributes, and gradually fuses them to integrate multiple levels of context. We evaluate our model and several baselines on TRANS, and set a new benchmark for the community to work on pedestrian stop and go forecasting.

3.Behavioural Curves Analysis Using Near-Infrared-Iris Image Sequences ⬇️

This paper proposes a new method to estimate behavioural curves from a stream of Near-Infra-Red (NIR) iris video frames. This method can be used in a Fitness For Duty system (FFD). The research focuses on determining the effect of external factors such as alcohol, drugs, and sleepiness on the Central Nervous System (CNS). The aim is to analyse how this behaviour is represented on iris and pupil movements and if it is possible to capture these changes with a standard NIR camera. The behaviour analysis showed essential differences in pupil and iris behaviour to classify the workers in "Fit" or "Unfit" conditions. The best results can distinguish subjects robustly under alcohol, drug consumption, and sleep conditions. The Multi-Layer-Perceptron and Gradient Boosted Machine reached the best results in all groups with an overall accuracy for Fit and Unfit classes of 74.0% and 75.5%, respectively. These results open a new application for iris capture devices.

4.The Familiarity Hypothesis: Explaining the Behavior of Deep Open Set Methods ⬇️

In many object recognition applications, the set of possible categories is an open set, and the deployed recognition system will encounter novel objects belonging to categories unseen during training. Detecting such "novel category" objects is usually formulated as an anomaly detection problem. Anomaly detection algorithms for feature-vector data identify anomalies as outliers, but outlier detection has not worked well in deep learning. Instead, methods based on the computed logits of visual object classifiers give state-of-the-art performance. This paper proposes the Familiarity Hypothesis that these methods succeed because they are detecting the absence of familiar learned features rather than the presence of novelty. The paper reviews evidence from the literature and presents additional evidence from our own experiments that provide strong support for this hypothesis. The paper concludes with a discussion of whether familiarity detection is an inevitable consequence of representation learning.

5.Didn't see that coming: a survey on non-verbal social human behavior forecasting ⬇️

Non-verbal social human behavior forecasting has increasingly attracted the interest of the research community in recent years. Its direct applications to human-robot interaction and socially-aware human motion generation make it a very attractive field. In this survey, we define the behavior forecasting problem for multiple interactive agents in a generic way that aims at unifying the fields of social signals prediction and human motion forecasting, traditionally separated. We hold that both problem formulations refer to the same conceptual problem, and identify many shared fundamental challenges: future stochasticity, context awareness, history exploitation, etc. We also propose a taxonomy that comprises methods published in the last 5 years in a very informative way and describes the current main concerns of the community with regard to this problem. In order to promote further research on this field, we also provide a summarised and friendly overview of audiovisual datasets featuring non-acted social interactions. Finally, we describe the most common metrics used in this task and their particular issues.

6.Real-Time Hybrid Mapping of Populated Indoor Scenes using a Low-Cost Monocular UAV ⬇️

Unmanned aerial vehicles (UAVs) have been used for many applications in recent years, from urban search and rescue, to agricultural surveying, to autonomous underground mine exploration. However, deploying UAVs in tight, indoor spaces, especially close to humans, remains a challenge. One solution, when limited payload is required, is to use micro-UAVs, which pose less risk to humans and typically cost less to replace after a crash. However, micro-UAVs can only carry a limited sensor suite, e.g. a monocular camera instead of a stereo pair or LiDAR, complicating tasks like dense mapping and markerless multi-person 3D human pose estimation, which are needed to operate in tight environments around people. Monocular approaches to such tasks exist, and dense monocular mapping approaches have been successfully deployed for UAV applications. However, despite many recent works on both marker-based and markerless multi-UAV single-person motion capture, markerless single-camera multi-person 3D human pose estimation remains a much earlier-stage technology, and we are not aware of existing attempts to deploy it in an aerial context. In this paper, we present what is thus, to our knowledge, the first system to perform simultaneous mapping and multi-person 3D human pose estimation from a monocular camera mounted on a single UAV. In particular, we show how to loosely couple state-of-the-art monocular depth estimation and monocular 3D human pose estimation approaches to reconstruct a hybrid map of a populated indoor scene in real time. We validate our component-level design choices via extensive experiments on the large-scale ScanNet and GTA-IM datasets. To evaluate our system-level performance, we also construct a new Oxford Hybrid Mapping dataset of populated indoor scenes.

7.SFPN: Synthetic FPN for Object Detection ⬇️

FPN (Feature Pyramid Network) has become a basic component of most SoTA one stage object detectors. Many previous studies have repeatedly proved that FPN can caputre better multi-scale feature maps to more precisely describe objects if they are with different sizes. However, for most backbones such VGG, ResNet, or DenseNet, the feature maps at each layer are downsized to their quarters due to the pooling operation or convolutions with stride 2. The gap of down-scaling-by-2 is large and makes its FPN not fuse the features smoothly. This paper proposes a new SFPN (Synthetic Fusion Pyramid Network) arichtecture which creates various synthetic layers between layers of the original FPN to enhance the accuracy of light-weight CNN backones to extract objects' visual features more accurately. Finally, experiments prove the SFPN architecture outperforms either the large backbone VGG16, ResNet50 or light-weight backbones such as MobilenetV2 based on AP score.

8.Rethinking Efficient Lane Detection via Curve Modeling ⬇️

This paper presents a novel parametric curve-based method for lane detection in RGB images. Unlike state-of-the-art segmentation-based and point detection-based methods that typically require heuristics to either decode predictions or formulate a large sum of anchors, the curve-based methods can learn holistic lane representations naturally. To handle the optimization difficulties of existing polynomial curve methods, we propose to exploit the parametric Bézier curve due to its ease of computation, stability, and high freedom degrees of transformations. In addition, we propose the deformable convolution-based feature flip fusion, for exploiting the symmetry properties of lanes in driving scenes. The proposed method achieves a new state-of-the-art performance on the popular LLAMAS benchmark. It also achieves favorable accuracy on the TuSimple and CULane datasets, while retaining both low latency (> 150 FPS) and small model size (< 10M). Our method can serve as a new baseline, to shed the light on the parametric curves modeling for lane detection. Codes of our model and PytorchAutoDrive: a unified framework for self-driving perception, are available at: this https URL .

9.DiT: Self-supervised Pre-training for Document Image Transformer ⬇️

Image Transformer has recently achieved significant progress for natural image understanding, either using supervised (ViT, DeiT, etc.) or self-supervised (BEiT, MAE, etc.) pre-training techniques. In this paper, we propose DiT, a self-supervised pre-trained Document Image Transformer model using large-scale unlabeled text images for Document AI tasks, which is essential since no supervised counterparts ever exist due to the lack of human labeled document images. We leverage DiT as the backbone network in a variety of vision-based Document AI tasks, including document image classification, document layout analysis, as well as table detection. Experiment results have illustrated that the self-supervised pre-trained DiT model achieves new state-of-the-art results on these downstream tasks, e.g. document image classification (91.11 $\rightarrow$ 92.69), document layout analysis (91.0 $\rightarrow$ 94.9) and table detection (94.23 $\rightarrow$ 96.55). The code and pre-trained models are publicly available at \url{this https URL}.

10.ViT-P: Rethinking Data-efficient Vision Transformers from Locality ⬇️

Recent advances of Transformers have brought new trust to computer vision tasks. However, on small dataset, Transformers is hard to train and has lower performance than convolutional neural networks. We make vision transformers as data-efficient as convolutional neural networks by introducing multi-focal attention bias. Inspired by the attention distance in a well-trained ViT, we constrain the self-attention of ViT to have multi-scale localized receptive field. The size of receptive field is adaptable during training so that optimal configuration can be learned. We provide empirical evidence that proper constrain of receptive field can reduce the amount of training data for vision transformers. On Cifar100, our ViT-P Base model achieves the state-of-the-art accuracy (83.16%) trained from scratch. We also perform analysis on ImageNet to show our method does not lose accuracy on large data sets.

11.Computer-Aided Road Inspection: Systems and Algorithms ⬇️

Road damage is an inconvenience and a safety hazard, severely affecting vehicle condition, driving comfort, and traffic safety. The traditional manual visual road inspection process is pricey, dangerous, exhausting, and cumbersome. Also, manual road inspection results are qualitative and subjective, as they depend entirely on the inspector's personal experience. Therefore, there is an ever-increasing need for automated road inspection systems. This chapter first compares the five most common road damage types. Then, 2-D/3-D road imaging systems are discussed. Finally, state-of-the-art machine vision and intelligence-based road damage detection algorithms are introduced.

12.F2DNet: Fast Focal Detection Network for Pedestrian Detection ⬇️

Two-stage detectors are state-of-the-art in object detection as well as pedestrian detection. However, the current two-stage detectors are inefficient as they do bounding box regression in multiple steps i.e. in region proposal networks and bounding box heads. Also, the anchor-based region proposal networks are computationally expensive to train. We propose F2DNet, a novel two-stage detection architecture which eliminates redundancy of current two-stage detectors by replacing the region proposal network with our focal detection network and bounding box head with our fast suppression head. We benchmark F2DNet on top pedestrian detection datasets, thoroughly compare it against the existing state-of-the-art detectors and conduct cross dataset evaluation to test the generalizability of our model to unseen data. Our F2DNet achieves 8.7%, 2.2%, and 6.1% MR-2 on City Persons, Caltech Pedestrian, and Euro City Person datasets respectively when trained on a single dataset and reaches 20.4% and 26.2% MR-2 in heavy occlusion setting of Caltech Pedestrian and City Persons datasets when using progressive fine-tunning. On top of that F2DNet have significantly lesser inference time compared to the current state-of-the-art. Code and trained models will be available at this https URL.

13.Quantum Levenberg--Marquardt Algorithm for optimization in Bundle Adjustment ⬇️

In this paper we develop a quantum optimization algorithm and use it to solve the bundle adjustment problem with a simulated quantum computer. Bundle adjustment is the process of optimizing camera poses and sensor properties to best reconstruct the three-dimensional structure and viewing parameters. This problem is often solved using some implementation of the Levenberg--Marquardt algorithm. In this case we implement a quantum algorithm for solving the linear system of normal equations that calculates the optimization step in Levenberg--Marquardt. This procedure is the current bottleneck in the algorithmic complexity of bundle adjustment. The proposed quantum algorithm dramatically reduces the complexity of this operation with respect to the number of points.
We investigate 9 configurations of a toy-model for bundle adjustment, limited to 10 points and 2 cameras. This optimization problem is solved both by using the sparse Levenberg-Marquardt algorithm and our quantum implementation. The resulting solutions are presented, showing an improved rate of convergence, together with an analysis of the theoretical speed up and the probability of running the algorithm successfully on a current quantum computer.
The presented quantum algorithm is a seminal implementation of using quantum computing algorithms in order to solve complex optimization problems in computer vision, in particular bundle adjustment, which offers several avenues of further investigations.

14.Mixed Reality Depth Contour Occlusion Using Binocular Similarity Matching and Three-dimensional Contour Optimisation ⬇️

Mixed reality applications often require virtual objects that are partly occluded by real objects. However, previous research and commercial products have limitations in terms of performance and efficiency. To address these challenges, we propose a novel depth contour occlusion (DCO) algorithm. The proposed method is based on the sensitivity of contour occlusion and a binocular stereoscopic vision device. In this method, a depth contour map is combined with a sparse depth map obtained from a two-stage adaptive filter area stereo matching algorithm and the depth contour information of the objects extracted by a digital image stabilisation optical flow method. We also propose a quadratic optimisation model with three constraints to generate an accurate dense map of the depth contour for high-quality real-virtual occlusion. The whole process is accelerated by GPU. To evaluate the effectiveness of the algorithm, we demonstrate a time con-sumption statistical analysis for each stage of the DCO algorithm execution. To verify the relia-bility of the real-virtual occlusion effect, we conduct an experimental analysis on single-sided, enclosed, and complex occlusions; subsequently, we compare it with the occlusion method without quadratic optimisation. With our GPU implementation for real-time DCO, the evaluation indicates that applying the presented DCO algorithm can enhance the real-time performance and the visual quality of real-virtual occlusion.

15.Freeform Body Motion Generation from Speech ⬇️

People naturally conduct spontaneous body motions to enhance their speeches while giving talks. Body motion generation from speech is inherently difficult due to the non-deterministic mapping from speech to body motions. Most existing works map speech to motion in a deterministic way by conditioning on certain styles, leading to sub-optimal results. Motivated by studies in linguistics, we decompose the co-speech motion into two complementary parts: pose modes and rhythmic dynamics. Accordingly, we introduce a novel freeform motion generation model (FreeMo) by equipping a two-stream architecture, i.e., a pose mode branch for primary posture generation, and a rhythmic motion branch for rhythmic dynamics synthesis. On one hand, diverse pose modes are generated by conditional sampling in a latent space, guided by speech semantics. On the other hand, rhythmic dynamics are synced with the speech prosody. Extensive experiments demonstrate the superior performance against several baselines, in terms of motion diversity, quality and syncing with speech. Code and pre-trained models will be publicly available through this https URL.

16.Semi-parametric Makeup Transfer via Semantic-aware Correspondence ⬇️

The large discrepancy between the source non-makeup image and the reference makeup image is one of the key challenges in makeup transfer. Conventional approaches for makeup transfer either learn disentangled representation or perform pixel-wise correspondence in a parametric way between two images. We argue that non-parametric techniques have a high potential for addressing the pose, expression, and occlusion discrepancies. To this end, this paper proposes a \textbf{S}emi-\textbf{p}arametric \textbf{M}akeup \textbf{T}ransfer (SpMT) method, which combines the reciprocal strengths of non-parametric and parametric mechanisms. The non-parametric component is a novel \textbf{S}emantic-\textbf{a}ware \textbf{C}orrespondence (SaC) module that explicitly reconstructs content representation with makeup representation under the strong constraint of component semantics. The reconstructed representation is desired to preserve the spatial and identity information of the source image while "wearing" the makeup of the reference image. The output image is synthesized via a parametric decoder that draws on the reconstructed representation. Extensive experiments demonstrate the superiority of our method in terms of visual quality, robustness, and flexibility. Code and pre-trained model are available at \url{this https URL.

17.Nuclei segmentation and classification in histopathology images with StarDist for the CoNIC Challenge 2022 ⬇️

Segmentation and classification of nuclei in histopathology images is an important task in computational pathology. Here we describe how we used StarDist, a deep learning based approach based on star-convex shape representations, for the Colon Nuclei Identification and Counting (CoNIC) challenge 2022.

18.A Comprehensive Review of Computer Vision in Sports: Open Issues, Future Trends and Research Directions ⬇️

Recent developments in video analysis of sports and computer vision techniques have achieved significant improvements to enable a variety of critical operations. To provide enhanced information, such as detailed complex analysis in sports like soccer, basketball, cricket, badminton, etc., studies have focused mainly on computer vision techniques employed to carry out different tasks. This paper presents a comprehensive review of sports video analysis for various applications high-level analysis such as detection and classification of players, tracking player or ball in sports and predicting the trajectories of player or ball, recognizing the teams strategies, classifying various events in sports. The paper further discusses published works in a variety of application-specific tasks related to sports and the present researchers views regarding them. Since there is a wide research scope in sports for deploying computer vision techniques in various sports, some of the publicly available datasets related to a particular sport have been provided. This work reviews a detailed discussion on some of the artificial intelligence(AI)applications in sports vision, GPU-based work stations, and embedded platforms. Finally, this review identifies the research directions, probable challenges, and future trends in the area of visual recognition in sports.

19.Feature Transformation for Cross-domain Few-shot Remote Sensing Scene Classification ⬇️

Effectively classifying remote sensing scenes is still a challenge due to the increasing spatial resolution of remote imaging and large variances between remote sensing images. Existing research has greatly improved the performance of remote sensing scene classification (RSSC). However, these methods are not applicable to cross-domain few-shot problems where target domain is with very limited training samples available and has a different data distribution from source domain. To improve the model's applicability, we propose the feature-wise transformation module (FTM) in this paper. FTM transfers the feature distribution learned on source domain to that of target domain by a very simple affine operation with negligible additional parameters. Moreover, FTM can be effectively learned on target domain in the case of few training data available and is agnostic to specific network structures. Experiments on RSSC and land-cover mapping tasks verified its capability to handle cross-domain few-shot problems. By comparison with directly finetuning, FTM achieves better performance and possesses better transferability and fine-grained discriminability. \textit{Code will be publicly available.}

20.Class-Aware Contrastive Semi-Supervised Learning ⬇️

Pseudo-label-based semi-supervised learning (SSL) has achieved great success on raw data utilization. However, its training procedure suffers from confirmation bias due to the noise contained in self-generated artificial labels. Moreover, the model's judgment becomes noisier in real-world applications with extensive out-of-distribution data. To address this issue, we propose a general method named Class-aware Contrastive Semi-Supervised Learning (CCSSL), which is a drop-in helper to improve the pseudo-label quality and enhance the model's robustness in the real-world setting. Rather than treating real-world data as a union set, our method separately handles reliable in-distribution data with class-wise clustering for blending into downstream tasks and noisy out-of-distribution data with image-wise contrastive for better generalization. Furthermore, by applying target re-weighting, we successfully emphasize clean label learning and simultaneously reduce noisy label learning. Despite its simplicity, our proposed CCSSL has significant performance improvements over the state-of-the-art SSL methods on the standard datasets CIFAR100 and STL10. On the real-world dataset Semi-iNat 2021, we improve FixMatch by 9.80% and CoMatch by 3.18%.

21.Patch Similarity Aware Data-Free Quantization for Vision Transformers ⬇️

Vision transformers have recently gained great success on various computer vision tasks; nevertheless, their high model complexity makes it challenging to deploy on resource-constrained devices. Quantization is an effective approach to reduce model complexity, and data-free quantization, which can address data privacy and security concerns during model deployment, has received widespread interest. Unfortunately, all existing methods, such as BN regularization, were designed for convolutional neural networks and cannot be applied to vision transformers with significantly different model architectures. In this paper, we propose PSAQ-ViT, a Patch Similarity Aware data-free Quantization framework for Vision Transformers, to enable the generation of "realistic" samples based on the vision transformer's unique properties for calibrating the quantization parameters. Specifically, we analyze the self-attention module's properties and reveal a general difference (patch similarity) in its processing of Gaussian noise and real images. The above insights guide us to design a relative value metric to optimize the Gaussian noise to approximate the real images, which are then utilized to calibrate the quantization parameters. Extensive experiments and ablation studies are conducted on various benchmarks to validate the effectiveness of PSAQ-ViT, which can even outperform the real-data-driven methods.

22.Detecting GAN-generated Images by Orthogonal Training of Multiple CNNs ⬇️

In the last few years, we have witnessed the rise of a series of deep learning methods to generate synthetic images that look extremely realistic. These techniques prove useful in the movie industry and for artistic purposes. However, they also prove dangerous if used to spread fake news or to generate fake online accounts. For this reason, detecting if an image is an actual photograph or has been synthetically generated is becoming an urgent necessity. This paper proposes a detector of synthetic images based on an ensemble of Convolutional Neural Networks (CNNs). We consider the problem of detecting images generated with techniques not available at training time. This is a common scenario, given that new image generators are published more and more frequently. To solve this issue, we leverage two main ideas: (i) CNNs should provide orthogonal results to better contribute to the ensemble; (ii) original images are better defined than synthetic ones, thus they should be better trusted at testing time. Experiments show that pursuing these two ideas improves the detector accuracy on NVIDIA's newly generated StyleGAN3 images, never used in training.

23.OPAL: Occlusion Pattern Aware Loss for Unsupervised Light Field Disparity Estimation ⬇️

Light field disparity estimation is an essential task in computer vision with various applications. Although supervised learning-based methods have achieved both higher accuracy and efficiency than traditional optimization-based methods, the dependency on ground-truth disparity for training limits the overall generalization performance not to say for real-world scenarios where the ground-truth disparity is hard to capture. In this paper, we argue that unsupervised methods can achieve comparable accuracy, but, more importantly, much higher generalization capacity and efficiency than supervised methods. Specifically, we present the Occlusion Pattern Aware Loss, named OPAL, which successfully extracts and encodes the general occlusion patterns inherent in the light field for loss calculation. OPAL enables i) accurate and robust estimation by effectively handling occlusions without using any ground-truth information for training and ii) much efficient performance by significantly reducing the network parameters required for accurate inference. Besides, a transformer-based network and a refinement module are proposed for achieving even more accurate results. Extensive experiments demonstrate our method not only significantly improves the accuracy compared with the SOTA unsupervised methods, but also possesses strong generalization capacity, even for real-world data, compared with supervised methods. Our code will be made publicly available.

24.Partial Wasserstein Adversarial Network for Non-rigid Point Set Registration ⬇️

Given two point sets, the problem of registration is to recover a transformation that matches one set to the other. This task is challenging due to the presence of the large number of outliers, the unknown non-rigid deformations and the large sizes of point sets. To obtain strong robustness against outliers, we formulate the registration problem as a partial distribution matching (PDM) problem, where the goal is to partially match the distributions represented by point sets in a metric space. To handle large point sets, we propose a scalable PDM algorithm by utilizing the efficient partial Wasserstein-1 (PW) discrepancy. Specifically, we derive the Kantorovich-Rubinstein duality for the PW discrepancy, and show its gradient can be explicitly computed. Based on these results, we propose a partial Wasserstein adversarial network (PWAN), which is able to approximate the PW discrepancy by a neural network, and minimize it by gradient descent. In addition, it also incorporates an efficient coherence regularizer for non-rigid transformations to avoid unrealistic deformations. We evaluate PWAN on practical point set registration tasks, and show that the proposed PWAN is robust, scalable and performs more favorably than the state-of-the-art methods.

25.Voice-Face Homogeneity Tells Deepfake ⬇️

Detecting forgery videos is highly desired due to the abuse of deepfake. Existing detection approaches contribute to exploring the specific artifacts in deepfake videos and fit well on certain data. However, the growing technique on these artifacts keeps challenging the robustness of traditional deepfake detectors. As a result, the development of generalizability of these approaches has reached a blockage. To address this issue, given the empirical results that the identities behind voices and faces are often mismatched in deepfake videos, and the voices and faces have homogeneity to some extent, in this paper, we propose to perform the deepfake detection from an unexplored voice-face matching view. To this end, a voice-face matching detection model is devised to measure the matching degree of these two on a generic audio-visual dataset. Thereafter, this model can be smoothly transferred to deepfake datasets without any fine-tuning, and the generalization across datasets is accordingly enhanced. We conduct extensive experiments over two widely exploited datasets - DFDC and FakeAVCeleb. Our model obtains significantly improved performance as compared to other state-of-the-art competitors and maintains favorable generalizability. The code has been released at this https URL.

26.Rethinking Reconstruction Autoencoder-Based Out-of-Distribution Detection ⬇️

In some scenarios, classifier requires detecting out-of-distribution samples far from its training data. With desirable characteristics, reconstruction autoencoder-based methods deal with this problem by using input reconstruction error as a metric of novelty vs. normality. We formulate the essence of such approach as a quadruplet domain translation with an intrinsic bias to only query for a proxy of conditional data uncertainty. Accordingly, an improvement direction is formalized as maximumly compressing the autoencoder's latent space while ensuring its reconstructive power for acting as a described domain translator. From it, strategies are introduced including semantic reconstruction, data certainty decomposition and normalized L2 distance to substantially improve original methods, which together establish state-of-the-art performance on various benchmarks, e.g., the FPR@95%TPR of CIFAR-100 vs. TinyImagenet-crop on Wide-ResNet is 0.2%. Importantly, our method works without any additional data, hard-to-implement structure, time-consuming pipeline, and even harming the classification accuracy of known classes.

27.Time-to-Label: Temporal Consistency for Self-Supervised Monocular 3D Object Detection ⬇️

Monocular 3D object detection continues to attract attention due to the cost benefits and wider availability of RGB cameras. Despite the recent advances and the ability to acquire data at scale, annotation cost and complexity still limit the size of 3D object detection datasets in the supervised settings. Self-supervised methods, on the other hand, aim at training deep networks relying on pretext tasks or various consistency constraints. Moreover, other 3D perception tasks (such as depth estimation) have shown the benefits of temporal priors as a self-supervision signal. In this work, we argue that the temporal consistency on the level of object poses, provides an important supervision signal given the strong prior on physical motion. Specifically, we propose a self-supervised loss which uses this consistency, in addition to render-and-compare losses, to refine noisy pose predictions and derive high-quality pseudo labels. To assess the effectiveness of the proposed method, we finetune a synthetically trained monocular 3D object detection model using the pseudo-labels that we generated on real data. Evaluation on the standard KITTI3D benchmark demonstrates that our method reaches competitive performance compared to other monocular self-supervised and supervised methods.

28.Semantic-Aware Representation Blending for Multi-Label Image Recognition with Partial Labels ⬇️

Training the multi-label image recognition models with partial labels, in which merely some labels are known while others are unknown for each image, is a considerably challenging and practical task. To address this task, current algorithms mainly depend on pre-training classification or similarity models to generate pseudo labels for the unknown labels. However, these algorithms depend on sufficient multi-label annotations to train the models, leading to poor performance especially with low known label proportion. In this work, we propose to blend category-specific representation across different images to transfer information of known labels to complement unknown labels, which can get rid of pre-training models and thus does not depend on sufficient annotations. To this end, we design a unified semantic-aware representation blending (SARB) framework that exploits instance-level and prototype-level semantic representation to complement unknown labels by two complementary modules: 1) an instance-level representation blending (ILRB) module blends the representations of the known labels in an image to the representations of the unknown labels in another image to complement these unknown labels. 2) a prototype-level representation blending (PLRB) module learns more stable representation prototypes for each category and blends the representation of unknown labels with the prototypes of corresponding labels to complement these labels. Extensive experiments on the MS-COCO, Visual Genome, Pascal VOC 2007 datasets show that the proposed SARB framework obtains superior performance over current leading competitors on all known label proportion settings, i.e., with the mAP improvement of 4.6%, 4.%, 2.2% on these three datasets when the known label proportion is 10%. Codes are available at this https URL.

29.DetFlowTrack: 3D Multi-object Tracking based on Simultaneous Optimization of Object Detection and Scene Flow Estimation ⬇️

3D Multi-Object Tracking (MOT) is an important part of the unmanned vehicle perception module. Most methods optimize object detection and data association independently. These methods make the network structure complicated and limit the improvement of MOT accuracy. we proposed a 3D MOT framework based on simultaneous optimization of object detection and scene flow estimation. In the framework, a detection-guidance scene flow module is proposed to relieve the problem of incorrect inter-frame assocation. For more accurate scene flow label especially in the case of motion with rotation, a box-transformation-based scene flow ground truth calculation method is proposed. Experimental results on the KITTI MOT dataset show competitive results over the state-of-the-arts and the robustness under extreme motion with rotation.

30.PatchMVSNet: Patch-wise Unsupervised Multi-View Stereo for Weakly-Textured Surface Reconstruction ⬇️

Learning-based multi-view stereo (MVS) has gained fine reconstructions on popular datasets. However, supervised learning methods require ground truth for training, which is hard to be collected, especially for the large-scale datasets. Though nowadays unsupervised learning methods have been proposed and have gotten gratifying results, those methods still fail to reconstruct intact results in challenging scenes, such as weakly-textured surfaces, as those methods primarily depend on pixel-wise photometric consistency which is subjected to various illuminations. To alleviate matching ambiguity in those challenging scenes, this paper proposes robust loss functions leveraging constraints beneath multi-view images: 1) Patch-wise photometric consistency loss, which expands the receptive field of the features in multi-view similarity measuring, 2) Robust twoview geometric consistency, which includes a cross-view depth consistency checking with the minimum occlusion. Our unsupervised strategy can be implemented with arbitrary depth estimation frameworks and can be trained with arbitrary large-scale MVS datasets. Experiments show that our method can decrease the matching ambiguity and particularly improve the completeness of weakly-textured reconstruction. Moreover, our method reaches the performance of the state-of-the-art methods on popular benchmarks, like DTU, Tanks and Temples and ETH3D. The code will be released soon.

31.ACVNet: Attention Concatenation Volume for Accurate and Efficient Stereo Matching ⬇️

Stereo matching is a fundamental building block for many vision and robotics applications. An informative and concise cost volume representation is vital for stereo matching of high accuracy and efficiency. In this paper, we present a novel cost volume construction method which generates attention weights from correlation clues to suppress redundant information and enhance matching-related information in the concatenation volume. To generate reliable attention weights, we propose multi-level adaptive patch matching to improve the distinctiveness of the matching cost at different disparities even for textureless regions. The proposed cost volume is named attention concatenation volume (ACV) which can be seamlessly embedded into most stereo matching networks, the resulting networks can use a more lightweight aggregation network and meanwhile achieve higher accuracy, e.g. using only 1/25 parameters of the aggregation network can achieve higher accuracy for GwcNet. Furthermore, we design a highly accurate network (ACVNet) based on our ACV, which achieves state-of-the-art performance on several benchmarks.

32.A Versatile Multi-View Framework for LiDAR-based 3D Object Detection with Guidance from Panoptic Segmentation ⬇️

3D object detection using LiDAR data is an indispensable component for autonomous driving systems. Yet, only a few LiDAR-based 3D object detection methods leverage segmentation information to further guide the detection process. In this paper, we propose a novel multi-task framework that jointly performs 3D object detection and panoptic segmentation. In our method, the 3D object detection backbone in Bird's-Eye-View (BEV) plane is augmented by the injection of Range-View (RV) feature maps from the 3D panoptic segmentation backbone. This enables the detection backbone to leverage multi-view information to address the shortcomings of each projection view. Furthermore, foreground semantic information is incorporated to ease the detection task by highlighting the locations of each object class in the feature maps. Finally, a new center density heatmap generated based on the instance-level information further guides the detection backbone by suggesting possible box center locations for objects. Our method works with any BEV-based 3D object detection method, and as shown by extensive experiments on the nuScenes dataset, it provides significant performance gains. Notably, the proposed method based on a single-stage CenterPoint 3D object detection network achieved state-of-the-art performance on nuScenes 3D Detection Benchmark with 67.3 NDS.

33.Towards Benchmarking and Evaluating Deepfake Detection ⬇️

Deepfake detection automatically recognizes the manipulated medias through the analysis of the difference between manipulated and non-altered videos. It is natural to ask which are the top performers among the existing deepfake detection approaches to identify promising research directions and provide practical guidance. Unfortunately, it's difficult to conduct a sound benchmarking comparison of existing detection approaches using the results in the literature because evaluation conditions are inconsistent across studies. Our objective is to establish a comprehensive and consistent benchmark, to develop a repeatable evaluation procedure, and to measure the performance of a range of detection approaches so that the results can be compared soundly. A challenging dataset consisting of the manipulated samples generated by more than 13 different methods has been collected, and 11 popular detection approaches (9 algorithms) from the existing literature have been implemented and evaluated with 6 fair-minded and practical evaluation metrics. Finally, 92 models have been trained and 644 experiments have been performed for the evaluation. The results along with the shared data and evaluation methodology constitute a benchmark for comparing deepfake detection approaches and measuring progress.

34.FS-COCO: Towards Understanding of Freehand Sketches of Common Objects in Context ⬇️

We advance sketch research to scenes with the first dataset of freehand scene sketches, FS-COCO. With practical applications in mind, we collect sketches that convey well scene content but can be sketched within a few minutes by a person with any sketching skills. Our dataset comprises 10,000 freehand scene vector sketches with per point space-time information by 100 non-expert individuals, offering both object- and scene-level abstraction. Each sketch is augmented with its text description. Using our dataset, we study for the first time the problem of the fine-grained image retrieval from freehand scene sketches and sketch captions. We draw insights on (i) Scene salience encoded in sketches with strokes temporal order; (ii) The retrieval performance accuracy from scene sketches against image captions; (iii) Complementarity of information in sketches and image captions, as well as the potential benefit of combining the two modalities. In addition, we propose new solutions enabled by our dataset (i) We adopt meta-learning to show how the retrieval model can be fine-tuned to a new user style given just a small set of sketches, (ii) We extend a popular vector sketch LSTM-based encoder to handle sketches with larger complexity than was supported by previous work. Namely, we propose a hierarchical sketch decoder, which we leverage at a sketch-specific "pretext" task. Our dataset enables for the first time research on freehand scene sketch understanding and its practical applications.

35.Pseudo-Stereo for Monocular 3D Object Detection in Autonomous Driving ⬇️

Pseudo-LiDAR 3D detectors have made remarkable progress in monocular 3D detection by enhancing the capability of perceiving depth with depth estimation networks, and using LiDAR-based 3D detection architectures. The advanced stereo 3D detectors can also accurately localize 3D objects. The gap in image-to-image generation for stereo views is much smaller than that in image-to-LiDAR generation. Motivated by this, we propose a Pseudo-Stereo 3D detection framework with three novel virtual view generation methods, including image-level generation, feature-level generation, and feature-clone, for detecting 3D objects from a single image. Our analysis of depth-aware learning shows that the depth loss is effective in only feature-level virtual view generation and the estimated depth map is effective in both image-level and feature-level in our framework. We propose a disparity-wise dynamic convolution with dynamic kernels sampled from the disparity feature map to filter the features adaptively from a single image for generating virtual image features, which eases the feature degradation caused by the depth estimation errors. Till submission (November 18, 2021), our Pseudo-Stereo 3D detection framework ranks 1st on car, pedestrian, and cyclist among the monocular 3D detectors with publications on the KITTI-3D benchmark. The code is released at this https URL.

36.Interactive Image Synthesis with Panoptic Layout Generation ⬇️

Interactive image synthesis from user-guided input is a challenging task when users wish to control the scene structure of a generated image with ease.Although remarkable progress has been made on layout-based image synthesis approaches, in order to get realistic fake image in interactive scene, existing methods require high-precision inputs, which probably need adjustment several times and are unfriendly to novice users. When placement of bounding boxes is subject to perturbation, layout-based models suffer from "missing regions" in the constructed semantic layouts and hence undesirable artifacts in the generated images. In this work, we propose Panoptic Layout Generative Adversarial Networks (PLGAN) to address this challenge. The PLGAN employs panoptic theory which distinguishes object categories between "stuff" with amorphous boundaries and "things" with well-defined shapes, such that stuff and instance layouts are constructed through separate branches and later fused into panoptic layouts. In particular, the stuff layouts can take amorphous shapes and fill up the missing regions left out by the instance layouts. We experimentally compare our PLGAN with state-of-the-art layout-based models on the COCO-Stuff, Visual Genome, and Landscape datasets. The advantages of PLGAN are not only visually demonstrated but quantitatively verified in terms of inception score, Fréchet inception distance, classification accuracy score, and coverage.

37.Sim2Real Instance-Level Style Transfer for 6D Pose Estimation ⬇️

In recent years, synthetic data has been widely used in the training of 6D pose estimation networks, in part because it automatically provides perfect annotation at low cost. However, there are still non-trivial domain gaps, such as differences in textures/materials, between synthetic and real data. These gaps have a measurable impact on performance. To solve this problem, we introduce a simulation to reality (sim2real) instance-level style transfer for 6D pose estimation network training. Our approach transfers the style of target objects individually, from synthetic to real, without human intervention. This improves the quality of synthetic data for training pose estimation networks. We also propose a complete pipeline from data collection to the training of a pose estimation network and conduct extensive evaluation on a real-world robotic platform. Our evaluation shows significant improvement achieved by our method in both pose estimation performance and the realism of images adapted by the style transfer.

38.Fast Neural Architecture Search for Lightweight Dense Prediction Networks ⬇️

We present LDP, a lightweight dense prediction neural architecture search (NAS) framework. Starting from a pre-defined generic backbone, LDP applies the novel Assisted Tabu Search for efficient architecture exploration. LDP is fast and suitable for various dense estimation problems, unlike previous NAS methods that are either computational demanding or deployed only for a single subtask. The performance of LPD is evaluated on monocular depth estimation, semantic segmentation, and image super-resolution tasks on diverse datasets, including NYU-Depth-v2, KITTI, Cityscapes, COCO-stuff, DIV2K, Set5, Set14, BSD100, Urban100. Experiments show that the proposed framework yields consistent improvements on all tested dense prediction tasks, while being $5%-315%$ more compact in terms of the number of model parameters than prior arts.

39.Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values ⬇️

We present Polarity Sampling, a theoretically justified plug-and-play method for controlling the generation quality and diversity of pre-trained deep generative networks DGNs). Leveraging the fact that DGNs are, or can be approximated by, continuous piecewise affine splines, we derive the analytical DGN output space distribution as a function of the product of the DGN's Jacobian singular values raised to a power $\rho$. We dub $\rho$ the $\textbf{polarity}$ parameter and prove that $\rho$ focuses the DGN sampling on the modes ($\rho &lt; 0$) or anti-modes ($\rho &gt; 0$) of the DGN output-space distribution. We demonstrate that nonzero polarity values achieve a better precision-recall (quality-diversity) Pareto frontier than standard methods, such as truncation, for a number of state-of-the-art DGNs. We also present quantitative and qualitative results on the improvement of overall generation quality (e.g., in terms of the Frechet Inception Distance) for a number of state-of-the-art DGNs, including StyleGAN3, BigGAN-deep, NVAE, for different conditional and unconditional image generation tasks. In particular, Polarity Sampling redefines the state-of-the-art for StyleGAN2 on the FFHQ Dataset to FID 2.57, StyleGAN2 on the LSUN Car Dataset to FID 2.27 and StyleGAN3 on the AFHQv2 Dataset to FID 3.95. Demo: this http URL

40.Towards Rich, Portable, and Large-Scale Pedestrian Data Collection ⬇️

Recently, pedestrian behavior research has shifted towards machine learning based methods and converged on the topic of modeling pedestrian interactions. For this, a large-scale dataset that contains rich information is needed. We propose a data collection system that is portable, which facilitates accessible large-scale data collection in diverse environments. We also couple the system with a semi-autonomous labeling pipeline for fast trajectory label production. We demonstrate the effectiveness of our system by further introducing a dataset we have collected -- the TBD pedestrian dataset. Compared with existing pedestrian datasets, our dataset contains three components: human verified labels grounded in the metric space, a combination of top-down and perspective views, and naturalistic human behavior in the presence of a socially appropriate "robot". In addition, the TBD pedestrian dataset is larger in quantity compared to similar existing datasets and contains unique pedestrian behavior.

41.Autoregressive Image Generation using Residual Quantization ⬇️

For autoregressive (AR) modeling of high-resolution images, vector quantization (VQ) represents an image as a sequence of discrete codes. A short sequence length is important for an AR model to reduce its computational costs to consider long-range interactions of codes. However, we postulate that previous VQ cannot shorten the code sequence and generate high-fidelity images together in terms of the rate-distortion trade-off. In this study, we propose the two-stage framework, which consists of Residual-Quantized VAE (RQ-VAE) and RQ-Transformer, to effectively generate high-resolution images. Given a fixed codebook size, RQ-VAE can precisely approximate a feature map of an image and represent the image as a stacked map of discrete codes. Then, RQ-Transformer learns to predict the quantized feature vector at the next position by predicting the next stack of codes. Thanks to the precise approximation of RQ-VAE, we can represent a 256$\times$256 image as 8$\times$8 resolution of the feature map, and RQ-Transformer can efficiently reduce the computational costs. Consequently, our framework outperforms the existing AR models on various benchmarks of unconditional and conditional image generation. Our approach also has a significantly faster sampling speed than previous AR models to generate high-quality images.

42.Contextformer: A Transformer with Spatio-Channel Attention for Context Modeling in Learned Image Compression ⬇️

Entropy modeling is a key component for high-performance image compression algorithms. Recent developments in autoregressive context modeling helped learning-based methods to surpass their classical counterparts. However, the performance of those models can be further improved due to the underexploited spatio-channel dependencies in latent space, and the suboptimal implementation of context adaptivity. Inspired by the adaptive characteristics of the transformers, we propose a transformer-based context model, a.k.a. Contextformer, which generalizes the de facto standard attention mechanism to spatio-channel attention. We replace the context model of a modern compression framework with the Contextformer and test it on the widely used Kodak image dataset. Our experimental results show that the proposed model provides up to 10% rate savings compared to the standard Versatile Video Coding (VVC) Test Model (VTM) 9.1, and outperforms various learning-based models.

43.Characterizing Renal Structures with 3D Block Aggregate Transformers ⬇️

Efficiently quantifying renal structures can provide distinct spatial context and facilitate biomarker discovery for kidney morphology. However, the development and evaluation of the transformer model to segment the renal cortex, medulla, and collecting system remains challenging due to data inefficiency. Inspired by the hierarchical structures in vision transformer, we propose a novel method using a 3D block aggregation transformer for segmenting kidney components on contrast-enhanced CT scans. We construct the first cohort of renal substructures segmentation dataset with 116 subjects under institutional review board (IRB) approval. Our method yields the state-of-the-art performance (Dice of 0.8467) against the baseline approach of 0.8308 with the data-efficient design. The Pearson R achieves 0.9891 between the proposed method and manual standards and indicates the strong correlation and reproducibility for volumetric analysis. We extend the proposed method to the public KiTS dataset, the method leads to improved accuracy compared to transformer-based approaches. We show that the 3D block aggregation transformer can achieve local communication between sequence representations without modifying self-attention, and it can serve as an accurate and efficient quantification tool for characterizing renal structures.

44.Differentiable Control Barrier Functions for Vision-based End-to-End Autonomous Driving ⬇️

Guaranteeing safety of perception-based learning systems is challenging due to the absence of ground-truth state information unlike in state-aware control scenarios. In this paper, we introduce a safety guaranteed learning framework for vision-based end-to-end autonomous driving. To this end, we design a learning system equipped with differentiable control barrier functions (dCBFs) that is trained end-to-end by gradient descent. Our models are composed of conventional neural network architectures and dCBFs. They are interpretable at scale, achieve great test performance under limited training data, and are safety guaranteed in a series of autonomous driving scenarios such as lane keeping and obstacle avoidance. We evaluated our framework in a sim-to-real environment, and tested on a real autonomous car, achieving safe lane following and obstacle avoidance via Augmented Reality (AR) and real parked vehicles.

45.Mobile authentication of copy detection patterns ⬇️

In the recent years, the copy detection patterns (CDP) attracted a lot of attention as a link between the physical and digital worlds, which is of great interest for the internet of things and brand protection applications. However, the security of CDP in terms of their reproducibility by unauthorized parties or clonability remains largely unexplored. In this respect this paper addresses a problem of anti-counterfeiting of physical objects and aims at investigating the authentication aspects and the resistances to illegal copying of the modern CDP from machine learning perspectives. A special attention is paid to a reliable authentication under the real life verification conditions when the codes are printed on an industrial printer and enrolled via modern mobile phones under regular light conditions. The theoretical and empirical investigation of authentication aspects of CDP is performed with respect to four types of copy fakes from the point of view of (i) multi-class supervised classification as a baseline approach and (ii) one-class classification as a real-life application case. The obtained results show that the modern machine-learning approaches and the technical capacities of modern mobile phones allow to reliably authenticate CDP on end-user mobile phones under the considered classes of fakes.

46.Simultaneous Alignment and Surface Regression Using Hybrid 2D-3D Networks for 3D Coherent Layer Segmentation of Retina OCT Images ⬇️

Automated surface segmentation of retinal layer is important and challenging in analyzing optical coherence tomography (OCT). Recently, many deep learning based methods have been developed for this task and yield remarkable performance. However, due to large spatial gap and potential mismatch between the B-scans of OCT data, all of them are based on 2D segmentation of individual B-scans, which may loss the continuity information across the B-scans. In addition, 3D surface of the retina layers can provide more diagnostic information, which is crucial in quantitative image analysis. In this study, a novel framework based on hybrid 2D-3D convolutional neural networks (CNNs) is proposed to obtain continuous 3D retinal layer surfaces from OCT. The 2D features of individual B-scans are extracted by an encoder consisting of 2D convolutions. These 2D features are then used to produce the alignment displacement field and layer segmentation by two 3D decoders, which are coupled via a spatial transformer module. The entire framework is trained end-to-end. To the best of our knowledge, this is the first study that attempts 3D retinal layer segmentation in volumetric OCT images based on CNNs. Experiments on a publicly available dataset show that our framework achieves superior results to state-of-the-art 2D methods in terms of both layer segmentation accuracy and cross-B-scan 3D continuity, thus offering more clinical values than previous works.

47.AutoMO-Mixer: An automated multi-objective Mixer model for balanced, safe and robust prediction in medicine ⬇️

Accurately identifying patient's status through medical images plays an important role in diagnosis and treatment. Artificial intelligence (AI), especially the deep learning, has achieved great success in many fields. However, more reliable AI model is needed in image guided diagnosis and therapy. To achieve this goal, developing a balanced, safe and robust model with a unified framework is desirable. In this study, a new unified model termed as automated multi-objective Mixer (AutoMO-Mixer) model was developed, which utilized a recent developed multiple layer perceptron Mixer (MLP-Mixer) as base. To build a balanced model, sensitivity and specificity were considered as the objective functions simultaneously in training stage. Meanwhile, a new evidential reasoning based on entropy was developed to achieve a safe and robust model in testing stage. The experiment on an optical coherence tomography dataset demonstrated that AutoMO-Mixer can obtain safer, more balanced, and robust results compared with MLP-Mixer and other available models.

48.Uncertainty Estimation for Heatmap-based Landmark Localization ⬇️

Automatic anatomical landmark localization has made great strides by leveraging deep learning methods in recent years. The ability to quantify the uncertainty of these predictions is a vital ingredient needed to see these methods adopted in clinical use, where it is imperative that erroneous predictions are caught and corrected. We propose Quantile Binning, a data-driven method to categorise predictions by uncertainty with estimated error bounds. This framework can be applied to any continuous uncertainty measure, allowing straightforward identification of the best subset of predictions with accompanying estimated error bounds. We facilitate easy comparison between uncertainty measures by constructing two evaluation metrics derived from Quantile Binning. We demonstrate this framework by comparing and contrasting three uncertainty measures (a baseline, the current gold standard, and a proposed method combining aspects of the two), across two datasets (one easy, one hard) and two heatmap-based landmark localization model paradigms (U-Net and patch-based). We conclude by illustrating how filtering out gross mispredictions caught in our Quantile Bins significantly improves the proportion of predictions under an acceptable error threshold, and offer recommendations on which uncertainty measure to use and how to use it.

49.Do Explanations Explain? Model Knows Best ⬇️

It is a mystery which input features contribute to a neural network's output. Various explanation (feature attribution) methods are proposed in the literature to shed light on the problem. One peculiar observation is that these explanations (attributions) point to different features as being important. The phenomenon raises the question, which explanation to trust? We propose a framework for evaluating the explanations using the neural network model itself. The framework leverages the network to generate input features that impose a particular behavior on the output. Using the generated features, we devise controlled experimental setups to evaluate whether an explanation method conforms to an axiom. Thus we propose an empirical framework for axiomatic evaluation of explanation methods. We evaluate well-known and promising explanation solutions using the proposed framework. The framework provides a toolset to reveal properties and drawbacks within existing and future explanation solutions.

50.Safety-aware metrics for object detectors in autonomous driving ⬇️

We argue that object detectors in the safety critical domain should prioritize detection of objects that are most likely to interfere with the actions of the autonomous actor. Especially, this applies to objects that can impact the actor's safety and reliability. In the context of autonomous driving, we propose new object detection metrics that reward the correct identification of objects that are most likely to interact with the subject vehicle (i.e., the actor), and that may affect its driving decision. To achieve this, we build a criticality model to reward the detection of the objects based on proximity, orientation, and relative velocity with respect to the subject vehicle. Then, we apply our model on the recent autonomous driving dataset nuScenes, and we compare eight different object detectors. Results show that, in several settings, object detectors that perform best according to the nuScenes ranking are not the preferable ones when the focus is shifted on safety and reliability.

51.Carbon Footprint of Selecting and Training Deep Learning Models for Medical Image Analysis ⬇️

The increasing energy consumption and carbon footprint of deep learning (DL) due to growing compute requirements has become a cause of concern. In this work, we focus on the carbon footprint of developing DL models for medical image analysis (MIA), where volumetric images of high spatial resolution are handled. In this study, we present and compare the features of four tools from literature to quantify the carbon footprint of DL. Using one of these tools we estimate the carbon footprint of medical image segmentation pipelines. We choose nnU-net as the proxy for a medical image segmentation pipeline and experiment on three common datasets. With our work we hope to inform on the increasing energy costs incurred by MIA. We discuss simple strategies to cut-down the environmental impact that can make model selection and training processes more efficient.

52.Convolutional Analysis Operator Learning by End-To-End Training of Iterative Neural Networks ⬇️

The concept of sparsity has been extensively applied for regularization in image reconstruction. Typically, sparsifying transforms are either pre-trained on ground-truth images or adaptively trained during the reconstruction. Thereby, learning algorithms are designed to minimize some target function which encodes the desired properties of the transform. However, this procedure ignores the subsequently employed reconstruction algorithm as well as the physical model which is responsible for the image formation process. Iterative neural networks - which contain the physical model - can overcome these issues. In this work, we demonstrate how convolutional sparsifying filters can be efficiently learned by end-to-end training of iterative neural networks. We evaluated our approach on a non-Cartesian 2D cardiac cine MRI example and show that the obtained filters are better suitable for the corresponding reconstruction algorithm than the ones obtained by decoupled pre-training.

53.MF-Hovernet: An Extension of Hovernet for Colon Nuclei Identification and Counting (CoNiC) Challenge ⬇️

Nuclei Identification and Counting is the most important morphological feature of cancers, especially in the colon. Many deep learning-based methods have been proposed to deal with this problem. In this work, we construct an extension of Hovernet for nuclei identification and counting to address the problem named MF-Hovernet. Our proposed model is the combination of multiple filer block to Hovernet architecture. The current result shows the efficiency of multiple filter block to improve the performance of the original Hovernet model.

54.Transformations in Learned Image Compression from a Communication Perspective ⬇️

In this paper, a unified transformation method in learned image compression(LIC) is proposed from the perspective of communication. Firstly, the quantization in LIC is considered as a generalized channel with additive uniform noise. Moreover, the LIC is interpreted as a particular communication system according to the consistency in structures and optimization objectives. Thus, the technology of communication systems can be applied to guide the design of modules in LIC. Furthermore, a unified transform method based on signal modulation (TSM) is defined. In the view of TSM, the existing transformation methods are mathematically reduced to a linear modulation. A series of transformation methods, e.g. TPM and TJM, are obtained by extending to nonlinear modulation. The experimental results on various datasets and backbone architectures verify that the effectiveness and robustness of the proposed method. More importantly, it further confirms the feasibility of guiding LIC design from a communication perspective. For example, when backbone architecture is hyperprior combining context model, our method achieves 3.52$%$ BD-rate reduction over GDN on Kodak dataset without increasing complexity.

55.HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imaging ⬇️

The rapid development of deep learning provides a better solution for the end-to-end reconstruction of hyperspectral image (HSI). However, existing learning-based methods have two major defects. Firstly, networks with self-attention usually sacrifice internal resolution to balance model performance against complexity, losing fine-grained high-resolution (HR) features. Secondly, even if the optimization focusing on spatial-spectral domain learning (SDL) converges to the ideal solution, there is still a significant visual difference between the reconstructed HSI and the truth. Therefore, we propose a high-resolution dual-domain learning network (HDNet) for HSI reconstruction. On the one hand, the proposed HR spatial-spectral attention module with its efficient feature fusion provides continuous and fine pixel-level features. On the other hand, frequency domain learning (FDL) is introduced for HSI reconstruction to narrow the frequency domain discrepancy. Dynamic FDL supervision forces the model to reconstruct fine-grained frequencies and compensate for excessive smoothing and distortion caused by pixel-level losses. The HR pixel-level attention and frequency-level refinement in our HDNet mutually promote HSI perceptual quality. Extensive quantitative and qualitative evaluation experiments show that our method achieves SOTA performance on simulated and real HSI datasets. Code and models will be released.

56.3D endoscopic depth estimation using 3D surface-aware constraints ⬇️

Robotic-assisted surgery allows surgeons to conduct precise surgical operations with stereo vision and flexible motor control. However, the lack of 3D spatial perception limits situational awareness during procedures and hinders mastering surgical skills in the narrow abdominal space. Depth estimation, as a representative perception task, is typically defined as an image reconstruction problem. In this work, we show that depth estimation can be reformed from a 3D surface perspective. We propose a loss function for depth estimation that integrates the surface-aware constraints, leading to a faster and better convergence with the valid information from spatial information. In addition, camera parameters are incorporated into the training pipeline to increase the control and transparency of the depth estimation. We also integrate a specularity removal module to recover more buried image information. Quantitative experimental results on endoscopic datasets and user studies with medical professionals demonstrate the effectiveness of our method.

57.MixCL: Pixel label matters to contrastive learning ⬇️

Contrastive learning and self-supervised techniques have gained prevalence in computer vision for the past few years. It is essential for medical image analysis, which is often notorious for its lack of annotations. Most existing self-supervised methods applied in natural imaging tasks focus on designing proxy tasks for unlabeled data. For example, contrastive learning is often based on the fact that an image and its transformed version share the same identity. However, pixel annotations contain much valuable information for medical image segmentation, which is largely ignored in contrastive learning. In this work, we propose a novel pre-training framework called Mixed Contrastive Learning (MixCL) that leverages both image identities and pixel labels for better modeling by maintaining identity consistency, label consistency, and reconstruction consistency together. Consequently, thus pre-trained model has more robust representations that characterize medical images. Extensive experiments demonstrate the effectiveness of the proposed method, improving the baseline by 5.28% and 14.12% in Dice coefficient when 5% labeled data of Spleen and 15% of BTVC are used in fine-tuning, respectively.

58.FairPrune: Achieving Fairness Through Pruning for Dermatological Disease Diagnosis ⬇️

Many works have shown that deep learning-based medical image classification models can exhibit bias toward certain demographic attributes like race, gender, and age. Existing bias mitigation methods primarily focus on learning debiased models, which may not necessarily guarantee all sensitive information can be removed and usually comes with considerable accuracy degradation on both privileged and unprivileged groups. To tackle this issue, we propose a method, FairPrune, that achieves fairness by pruning. Conventionally, pruning is used to reduce the model size for efficient inference. However, we show that pruning can also be a powerful tool to achieve fairness. Our observation is that during pruning, each parameter in the model has different importance for different groups' accuracy. By pruning the parameters based on this importance difference, we can reduce the accuracy difference between the privileged group and the unprivileged group to improve fairness without a large accuracy drop. To this end, we use the second derivative of the parameters of a pre-trained model to quantify the importance of each parameter with respect to the model accuracy for each group. Experiments on two skin lesion diagnosis datasets over multiple sensitive attributes demonstrate that our method can greatly improve fairness while keeping the average accuracy of both groups as high as possible.

59.Learning Category-Level Generalizable Object Manipulation Policy via Generative Adversarial Self-Imitation Learning from Demonstrations ⬇️

Generalizable object manipulation skills are critical for intelligent and multi-functional robots to work in real-world complex scenes. Despite the recent progress in reinforcement learning, it is still very challenging to learn a generalizable manipulation policy that can handle a category of geometrically diverse articulated objects. In this work, we tackle this category-level object manipulation policy learning problem via imitation learning in a task-agnostic manner, where we assume no handcrafted dense rewards but only a terminal reward. Given this novel and challenging generalizable policy learning problem, we identify several key issues that can fail the previous imitation learning algorithms and hinder the generalization to unseen instances. We then propose several general but critical techniques, including generative adversarial self-imitation learning from demonstrations, progressive growing of discriminator, and instance-balancing for expert buffer, that accurately pinpoints and tackles these issues and can benefit category-level manipulation policy learning regardless of the tasks. Our experiments on ManiSkill benchmarks demonstrate a remarkable improvement on all tasks and our ablation studies further validate the contribution of each proposed technique.

60.Scribble-Supervised Medical Image Segmentation via Dual-Branch Network and Dynamically Mixed Pseudo Labels Supervision ⬇️

Medical image segmentation plays an irreplaceable role in computer-assisted diagnosis, treatment planning, and following-up. Collecting and annotating a large-scale dataset is crucial to training a powerful segmentation model, but producing high-quality segmentation masks is an expensive and time-consuming procedure. Recently, weakly-supervised learning that uses sparse annotations (points, scribbles, bounding boxes) for network training has achieved encouraging performance and shown the potential for annotation cost reduction. However, due to the limited supervision signal of sparse annotations, it is still challenging to employ them for networks training directly. In this work, we propose a simple yet efficient scribble-supervised image segmentation method and apply it to cardiac MRI segmentation. Specifically, we employ a dual-branch network with one encoder and two slightly different decoders for image segmentation and dynamically mix the two decoders' predictions to generate pseudo labels for auxiliary supervision. By combining the scribble supervision and auxiliary pseudo labels supervision, the dual-branch network can efficiently learn from scribble annotations end-to-end. Experiments on the public ACDC dataset show that our method performs better than current scribble-supervised segmentation methods and also outperforms several semi-supervised segmentation methods.

61.Learning Incrementally to Segment Multiple Organs in a CT Image ⬇️

There exists a large number of datasets for organ segmentation, which are partially annotated and sequentially constructed. A typical dataset is constructed at a certain time by curating medical images and annotating the organs of interest. In other words, new datasets with annotations of new organ categories are built over time. To unleash the potential behind these partially labeled, sequentially-constructed datasets, we propose to incrementally learn a multi-organ segmentation model. In each incremental learning (IL) stage, we lose the access to previous data and annotations, whose knowledge is assumingly captured by the current model, and gain the access to a new dataset with annotations of new organ categories, from which we learn to update the organ segmentation model to include the new organs. While IL is notorious for its `catastrophic forgetting' weakness in the context of natural image analysis, we experimentally discover that such a weakness mostly disappears for CT multi-organ segmentation. To further stabilize the model performance across the IL stages, we introduce a light memory module and some loss functions to restrain the representation of different categories in feature space, aggregating feature representation of the same class and separating feature representation of different classes. Extensive experiments on five open-sourced datasets are conducted to illustrate the effectiveness of our method.

62.Universal Segmentation of 33 Anatomies ⬇️

In the paper, we present an approach for learning a single model that universally segments 33 anatomical structures, including vertebrae, pelvic bones, and abdominal organs. Our model building has to address the following challenges. Firstly, while it is ideal to learn such a model from a large-scale, fully-annotated dataset, it is practically hard to curate such a dataset. Thus, we resort to learn from a union of multiple datasets, with each dataset containing the images that are partially labeled. Secondly, along the line of partial labelling, we contribute an open-source, large-scale vertebra segmentation dataset for the benefit of spine analysis community, CTSpine1K, boasting over 1,000 3D volumes and over 11K annotated vertebrae. Thirdly, in a 3D medical image segmentation task, due to the limitation of GPU memory, we always train a model using cropped patches as inputs instead a whole 3D volume, which limits the amount of contextual information to be learned. To this, we propose a cross-patch transformer module to fuse more information in adjacent patches, which enlarges the aggregated receptive field for improved segmentation performance. This is especially important for segmenting, say, the elongated spine. Based on 7 partially labeled datasets that collectively contain about 2,800 3D volumes, we successfully learn such a universal model. Finally, we evaluate the universal model on multiple open-source datasets, proving that our model has a good generalization performance and can potentially serve as a solid foundation for downstream tasks.

63.WPNAS: Neural Architecture Search by jointly using Weight Sharing and Predictor ⬇️

Weight sharing based and predictor based methods are two major types of fast neural architecture search methods. In this paper, we propose to jointly use weight sharing and predictor in a unified framework. First, we construct a SuperNet in a weight-sharing way and probabilisticly sample architectures from the SuperNet. To increase the correctness of the evaluation of architectures, besides direct evaluation using the inherited weights, we further apply a few-shot predictor to assess the architecture on the other hand. The final evaluation of the architecture is the combination of direct evaluation, the prediction from the predictor and the cost of the architecture. We regard the evaluation as a reward and apply a self-critical policy gradient approach to update the architecture probabilities. To further reduce the side effects of weight sharing, we propose a weakly weight sharing method by introducing another HyperNet. We conduct experiments on datasets including CIFAR-10, CIFAR-100 and ImageNet under NATS-Bench, DARTS and MobileNet search space. The proposed WPNAS method achieves state-of-the-art performance on these datasets.

64.An Efficient Subpopulation-based Membership Inference Attack ⬇️

Membership inference attacks allow a malicious entity to predict whether a sample is used during training of a victim model or not. State-of-the-art membership inference attacks have shown to achieve good accuracy which poses a great privacy threat. However, majority of SOTA attacks require training dozens to hundreds of shadow models to accurately infer membership. This huge computation cost raises questions about practicality of these attacks on deep models. In this paper, we introduce a fundamentally different MI attack approach which obviates the need to train hundreds of shadow models. Simply put, we compare the victim model output on the target sample versus the samples from the same subpopulation (i.e., semantically similar samples), instead of comparing it with the output of hundreds of shadow models. The intuition is that the model response should not be significantly different between the target sample and its subpopulation if it was not a training sample. In cases where subpopulation samples are not available to the attacker, we show that training only a single generative model can fulfill the requirement. Hence, we achieve the state-of-the-art membership inference accuracy while significantly reducing the training computation cost.

65.Mind the Gap: Understanding the Modality Gap in Multi-modal Contrastive Representation Learning ⬇️

We present modality gap, an intriguing geometric phenomenon of the representation space of multi-modal models. Specifically, we show that different data modalities (e.g. images and text) are embedded at arm's length in their shared representation in multi-modal models such as CLIP. Our systematic analysis demonstrates that this gap is caused by a combination of model initialization and contrastive learning optimization. In model initialization, we show empirically and theoretically that the representation of a common deep neural network is restricted to a narrow cone. As a consequence, in a multi-modal model with two encoders, the representations of the two modalities are clearly apart when the model is initialized. During optimization, contrastive learning keeps the different modalities separate by a certain distance, which is influenced by the temperature parameter in the loss function. Our experiments further demonstrate that varying the modality gap distance has a significant impact in improving the model's downstream zero-shot classification performance and fairness. Our code and data are available at this https URL

66.Anomaly Detection-Inspired Few-Shot Medical Image Segmentation Through Self-Supervision With Supervoxels ⬇️

Recent work has shown that label-efficient few-shot learning through self-supervision can achieve promising medical image segmentation results. However, few-shot segmentation models typically rely on prototype representations of the semantic classes, resulting in a loss of local information that can degrade performance. This is particularly problematic for the typically large and highly heterogeneous background class in medical image segmentation problems. Previous works have attempted to address this issue by learning additional prototypes for each class, but since the prototypes are based on a limited number of slices, we argue that this ad-hoc solution is insufficient to capture the background properties. Motivated by this, and the observation that the foreground class (e.g., one organ) is relatively homogeneous, we propose a novel anomaly detection-inspired approach to few-shot medical image segmentation in which we refrain from modeling the background explicitly. Instead, we rely solely on a single foreground prototype to compute anomaly scores for all query pixels. The segmentation is then performed by thresholding these anomaly scores using a learned threshold. Assisted by a novel self-supervision task that exploits the 3D structure of medical images through supervoxels, our proposed anomaly detection-inspired few-shot medical image segmentation model outperforms previous state-of-the-art approaches on two representative MRI datasets for the tasks of abdominal organ segmentation and cardiac segmentation.

67.DIME: Fine-grained Interpretations of Multimodal Models via Disentangled Local Explanations ⬇️

The ability for a human to understand an Artificial Intelligence (AI) model's decision-making process is critical in enabling stakeholders to visualize model behavior, perform model debugging, promote trust in AI models, and assist in collaborative human-AI decision-making. As a result, the research fields of interpretable and explainable AI have gained traction within AI communities as well as interdisciplinary scientists seeking to apply AI in their subject areas. In this paper, we focus on advancing the state-of-the-art in interpreting multimodal models - a class of machine learning methods that tackle core challenges in representing and capturing interactions between heterogeneous data sources such as images, text, audio, and time-series data. Multimodal models have proliferated numerous real-world applications across healthcare, robotics, multimedia, affective computing, and human-computer interaction. By performing model disentanglement into unimodal contributions (UC) and multimodal interactions (MI), our proposed approach, DIME, enables accurate and fine-grained analysis of multimodal models while maintaining generality across arbitrary modalities, model architectures, and tasks. Through a comprehensive suite of experiments on both synthetic and real-world multimodal tasks, we show that DIME generates accurate disentangled explanations, helps users of multimodal models gain a deeper understanding of model behavior, and presents a step towards debugging and improving these models for real-world deployment. Code for our experiments can be found at this https URL.

68.Why adversarial training can hurt robust accuracy ⬇️

Machine learning classifiers with high test accuracy often perform poorly under adversarial attacks. It is commonly believed that adversarial training alleviates this issue. In this paper, we demonstrate that, surprisingly, the opposite may be true -- Even though adversarial training helps when enough data is available, it may hurt robust generalization in the small sample size regime. We first prove this phenomenon for a high-dimensional linear classification setting with noiseless observations. Our proof provides explanatory insights that may also transfer to feature learning models. Further, we observe in experiments on standard image datasets that the same behavior occurs for perceptible attacks that effectively reduce class information such as mask attacks and object corruptions.

69.Counting Molecules: Python based scheme for automated enumeration and categorization of molecules in scanning tunneling microscopy images ⬇️

Scanning tunneling and atomic force microscopies (STM/nc-AFM) are rapidly progressing to offer unprecedented spatial resolution of a diverse array of chemical species. In particular, they are employed to characterize on-surface chemical reactions by directly examining precursors and products. Chiral effects and self-assembled structures can also be investigated. This open source, modular, python based scheme automates the categorization of a variety of molecules present in medium sized (10$\times$10 to 100$\times$100 nm) scanned probe images.

70.Region-of-Interest Based Neural Video Compression ⬇️

Humans do not perceive all parts of a scene with the same resolution, but rather focus on few regions of interest (ROIs). Traditional Object-Based codecs take advantage of this biological intuition, and are capable of non-uniform allocation of bits in favor of salient regions, at the expense of increased distortion the remaining areas: such a strategy allows a boost in perceptual quality under low rate constraints. Recently, several neural codecs have been introduced for video compression, yet they operate uniformly over all spatial locations, lacking the capability of ROI-based processing. In this paper, we introduce two models for ROI-based neural video coding. First, we propose an implicit model that is fed with a binary ROI mask and it is trained by de-emphasizing the distortion of the background. Secondly, we design an explicit latent scaling method, that allows control over the quantization binwidth for different spatial regions of latent variables, conditioned on the ROI mask. By extensive experiments, we show that our methods outperform all our baselines in terms of Rate-Distortion (R-D) performance in the ROI. Moreover, they can generalize to different datasets and to any arbitrary ROI at inference time. Finally, they do not require expensive pixel-level annotations during training, as synthetic ROI masks can be used with little to no degradation in performance. To the best of our knowledge, our proposals are the first solutions that integrate ROI-based capabilities into neural video compression models.

71.Audio-Visual Object Classification for Human-Robot Collaboration ⬇️

Human-robot collaboration requires the contactless estimation of the physical properties of containers manipulated by a person, for example while pouring content in a cup or moving a food box. Acoustic and visual signals can be used to estimate the physical properties of such objects, which may vary substantially in shape, material and size, and also be occluded by the hands of the person. To facilitate comparisons and stimulate progress in solving this problem, we present the CORSMAL challenge and a dataset to assess the performance of the algorithms through a set of well-defined performance scores. The tasks of the challenge are the estimation of the mass, capacity, and dimensions of the object (container), and the classification of the type and amount of its content. A novel feature of the challenge is our real-to-simulation framework for visualising and assessing the impact of estimation errors in human-to-robot handovers.

72.Robust Segmentation of Brain MRI in the Wild with Hierarchical CNNs and no Retraining ⬇️

Retrospective analysis of brain MRI scans acquired in the clinic has the potential to enable neuroimaging studies with sample sizes much larger than those found in research datasets. However, analysing such clinical images "in the wild" is challenging, since subjects are scanned with highly variable protocols (MR contrast, resolution, orientation, etc.). Nevertheless, recent advances in convolutional neural networks (CNNs) and domain randomisation for image segmentation, best represented by the publicly available method SynthSeg, may enable morphometry of clinical MRI at scale. In this work, we first evaluate SynthSeg on an uncurated, heterogeneous dataset of more than 10,000 scans acquired at Massachusetts General Hospital. We show that SynthSeg is generally robust, but frequently falters on scans with low signal-to-noise ratio or poor tissue contrast. Next, we propose SynthSeg+, a novel method that greatly mitigates these problems using a hierarchy of conditional segmentation and denoising CNNs. We show that this method is considerably more robust than SynthSeg, while also outperforming cascaded networks and state-of-the-art segmentation denoising methods. Finally, we apply our approach to a proof-of-concept volumetric study of ageing, where it closely replicates atrophy patterns observed in research studies conducted on high-quality, 1mm, T1-weighted scans. The code and trained model are publicly available at this https URL.

73.A multi-stream convolutional neural network for classification of progressive MCI in Alzheimer's disease using structural MRI images ⬇️

Early diagnosis of Alzheimer's disease and its prodromal stage, also known as mild cognitive impairment (MCI), is critical since some patients with progressive MCI will develop the disease. We propose a multi-stream deep convolutional neural network fed with patch-based imaging data to classify stable MCI and progressive MCI. First, we compare MRI images of Alzheimer's disease with cognitively normal subjects to identify distinct anatomical landmarks using a multivariate statistical test. These landmarks are then used to extract patches that are fed into the proposed multi-stream convolutional neural network to classify MRI images. Next, we train the architecture in a separate scenario using samples from Alzheimer's disease images, which are anatomically similar to the progressive MCI ones and cognitively normal images to compensate for the lack of progressive MCI training data. Finally, we transfer the trained model weights to the proposed architecture in order to fine-tune the model using progressive MCI and stable MCI data. Experimental results on the ADNI-1 dataset indicate that our method outperforms existing methods for MCI classification, with an F1-score of 85.96%.

74.Color Space-based HoVer-Net for Nuclei Instance Segmentation and Classification ⬇️

Nuclei segmentation and classification is the first and most crucial step that is utilized for many different microscopy medical analysis applications. However, it suffers from many issues such as the segmentation of small objects, imbalance, and fine-grained differences between types of nuclei. In this paper, multiple different contributions were done tackling these problems present. Firstly, the recently released "ConvNeXt" was used as the encoder for HoVer-Net model since it leverages the key components of transformers that make them perform well. Secondly, to enhance the visual differences between nuclei, a multi-channel color space-based approach is used to aid the model in extracting distinguishing features. Thirdly, Unified Focal loss (UFL) was used to tackle the background-foreground imbalance. Finally, Sharpness-Aware Minimization (SAM) was used to ensure generalizability of the model. Overall, we were able to outperform the current state-of-the-art (SOTA), HoVer-Net, on the preliminary test set of the CoNiC Challenge 2022 by 12.489% mPQ+.

75.Semantic-guided Image Virtual Attribute Learning for Noisy Multi-label Chest X-ray Classification ⬇️

Deep learning methods have shown outstanding classification accuracy in medical image analysis problems, which is largely attributed to the availability of large datasets manually annotated with clean labels. However, such manual annotation can be expensive to obtain for large datasets, so we may rely on machine-generated noisy labels. Many Chest X-ray (CXR) classifiers are modelled from datasets with machine-generated labels, but their training procedure is in general not robust to the presence of noisy-label samples and can overfit those samples to produce sub-optimal solutions. Furthermore, CXR datasets are mostly multi-label, so current noisy-label learning methods designed for multi-class problems cannot be easily adapted. To address such noisy multi-label CXR learning problem, we propose a new learning method based on estimating image virtual attributes using semantic information from the label to assist in the identification and correction of noisy multi-labels from training samples. Our experiments on diverse noisy multi-label training sets and clean testing sets show that our model has state-of-the-art accuracy and robustness across all datasets.

76.E-CIR: Event-Enhanced Continuous Intensity Recovery ⬇️

A camera begins to sense light the moment we press the shutter button. During the exposure interval, relative motion between the scene and the camera causes motion blur, a common undesirable visual artifact. This paper presents E-CIR, which converts a blurry image into a sharp video represented as a parametric function from time to intensity. E-CIR leverages events as an auxiliary input. We discuss how to exploit the temporal event structure to construct the parametric bases. We demonstrate how to train a deep learning model to predict the function coefficients. To improve the appearance consistency, we further introduce a refinement module to propagate visual features among consecutive frames. Compared to state-of-the-art event-enhanced deblurring approaches, E-CIR generates smoother and more realistic results. The implementation of E-CIR is available at this https URL.

77.Quality or Quantity: Toward a Unified Approach for Multi-organ Segmentation in Body CT ⬇️

Organ segmentation of medical images is a key step in virtual imaging trials. However, organ segmentation datasets are limited in terms of quality (because labels cover only a few organs) and quantity (since case numbers are limited). In this study, we explored the tradeoffs between quality and quantity. Our goal is to create a unified approach for multi-organ segmentation of body CT, which will facilitate the creation of large numbers of accurate virtual phantoms. Initially, we compared two segmentation architectures, 3D-Unet and DenseVNet, which were trained using XCAT data that is fully labeled with 22 organs, and chose the 3D-Unet as the better performing model. We used the XCAT-trained model to generate pseudo-labels for the CT-ORG dataset that has only 7 organs segmented. We performed two experiments: First, we trained 3D-UNet model on the XCAT dataset, representing quality data, and tested it on both XCAT and CT-ORG datasets. Second, we trained 3D-UNet after including the CT-ORG dataset into the training set to have more quantity. Performance improved for segmentation in the organs where we have true labels in both datasets and degraded when relying on pseudo-labels. When organs were labeled in both datasets, Exp-2 improved Average DSC in XCAT and CT-ORG by 1. This demonstrates that quality data is the key to improving the model's performance.

78.Temporal Context Matters: Enhancing Single Image Prediction with Disease Progression Representations ⬇️

Clinical outcome or severity prediction from medical images has largely focused on learning representations from single-timepoint or snapshot scans. It has been shown that disease progression can be better characterized by temporal imaging. We therefore hypothesized that outcome predictions can be improved by utilizing the disease progression information from sequential images. We present a deep learning approach that leverages temporal progression information to improve clinical outcome predictions from single-timepoint images. In our method, a self-attention based Temporal Convolutional Network (TCN) is used to learn a representation that is most reflective of the disease trajectory. Meanwhile, a Vision Transformer is pretrained in a self-supervised fashion to extract features from single-timepoint images. The key contribution is to design a recalibration module that employs maximum mean discrepancy loss (MMD) to align distributions of the above two contextual representations. We train our system to predict clinical outcomes and severity grades from single-timepoint images. Experiments on chest and osteoarthritis radiography datasets demonstrate that our approach outperforms other state-of-the-art techniques.

79.Contextual Attention Network: Transformer Meets U-Net ⬇️

Currently, convolutional neural networks (CNN) (e.g., U-Net) have become the de facto standard and attained immense success in medical image segmentation. However, as a downside, CNN based methods are a double-edged sword as they fail to build long-range dependencies and global context connections due to the limited receptive field that stems from the intrinsic characteristics of the convolution operation. Hence, recent articles have exploited Transformer variants for medical image segmentation tasks which open up great opportunities due to their innate capability of capturing long-range correlations through the attention mechanism. Although being feasibly designed, most of the cohort studies incur prohibitive performance in capturing local information, thereby resulting in less lucidness of boundary areas. In this paper, we propose a contextual attention network to tackle the aforementioned limitations. The proposed method uses the strength of the Transformer module to model the long-range contextual dependency. Simultaneously, it utilizes the CNN encoder to capture local semantic information. In addition, an object-level representation is included to model the regional interaction map. The extracted hierarchical features are then fed to the contextual attention module to adaptively recalibrate the representation space using the local information. Then, they emphasize the informative regions while taking into account the long-range contextual dependency derived by the Transformer module. We validate our method on several large-scale public medical image segmentation datasets and achieve state-of-the-art performance. We have provided the implementation code in this https URL.