Skip to content

[NeurIPS 2024 Spotlight] Motion Forecasting in Continuous Driving

Notifications You must be signed in to change notification settings

Zbozhou/RealMotion

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Motion Forecasting in Continuous Driving

Motion Forecasting in Continuous Driving
Nan Song, Bozhou Zhang, Xiatian Zhu, Li Zhang
NeurIPS 2024

🚗 Abstract

Motion forecasting for agents in autonomous driving is highly challenging due to the numerous possibilities for each agent's next action and their complex interactions in space and time. In real applications, motion forecasting takes place repeatedly and continuously as the self-driving car moves. However, existing forecasting methods typically process each driving scene within a certain range independently, totally ignoring the situational and contextual relationships between successive driving scenes. This significantly simplifies the forecasting task, making the solutions suboptimal and inefficient to use in practice. To address this fundamental limitation, we propose a novel motion forecasting framework for continuous driving, named RealMotion. It comprises two integral streams both at the scene level: (1) The scene context stream progressively accumulates historical scene information until the present moment, capturing temporal interactive relationships among scene elements. (2) The agent trajectory stream optimizes current forecasting by sequentially relaying past predictions. Besides, a data reorganization strategy is introduced to narrow the gap between existing benchmarks and real-world applications, consistent with our network. These approaches enable exploiting more broadly the situational and progressive insights of dynamic motion across space and time. Extensive experiments on Argoverse series with different settings demonstrate that our RealMotion achieves state-of-the-art performance, along with the advantage of efficient real-world inference.

🎞️ Pipeline


🛠️ Get started

Set up a new virtual environment

conda create -n motion python=3.10
conda activate motion

Install dependency packpages

pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cu118
pip install -r requirements.txt
pip install natten==0.17.3+torch200cu118 -f https://shi-labs.com/natten/wheels

🕹️ Prepare the data

data
    ├── train
    │   ├── 0000b0f9-99f9-4a1f-a231-5be9e4c523f7
    │   ├── 0000b6ab-e100-4f6b-aee8-b520b57c0530
    │   ├── ...
    ├── val
    │   ├── 00010486-9a07-48ae-b493-cf4545855937
    │   ├── 00062a32-8d6d-4449-9948-6fedac67bfcd
    │   ├── ...
    ├── test
    │   ├── 0000b329-f890-4c2b-93f2-7e2413d4ca5b
    │   ├── 0008c251-e9b0-4708-b762-b15cb6effc27
    │   ├── ...

Preprocess

python preprocess.py -d /path/to/data -p

The structure of the dataset after processing

└── data
    └── realmotion_processed
        ├── train
        ├── val
        └── test

🔥 Training and testing

# Train
python train.py

# Val
python eval.py checkpoint=/path/to/ckpt

# Test for submission
python eval.py checkpoint=/path/to/ckpt submit=true

⭐ Results and checkpoints

Models minADE1 minFDE1 minADE6 minFDE6 b-minFDE6 Checkpoint
RealMotion-I 1.808 4.510 0.728 1.420 2.043 remotion_i.pth
RealMotion 1.646 4.100 0.669 1.303 1.935 remotion.pth

📜 BibTeX

@inproceedings{song2024realmotion,
 title={Motion Forecasting in Continuous Driving},
 author={Song, Nan and Zhang, Bozhou and Zhu, Xiatian and Zhang, Li},
 booktitle={NeurIPS},
 year={2024},
}

❤️ Acknowledgements

About

[NeurIPS 2024 Spotlight] Motion Forecasting in Continuous Driving

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%