BlockGCN: Redefining Topology Awareness for Skeleton-Based Action Recognition

Performance vs. Model Size on NTU RGB+D 120 Cross-Subject Benchmark

Figure 1: This graph demonstrates the superior performance of BlockGCN compared to existing methods on the NTU RGB+D 120 Cross-Subject Benchmark. BlockGCN achieves higher accuracy with fewer parameters, indicating its efficiency and effectiveness.

Illustration of BlockGC

Figure 2: An illustration of the BlockGC structure within BlockGCN. BlockGC divides the feature dimension into multiple groups, applying spatial aggregation and feature projection in parallel to efficiently model high-level semantics.

Preparation

Install torchlight

Run pip install -e torchlight

Download datasets.

There are 3 datasets to download:

• NTU RGB+D 60 Skeleton
• NTU RGB+D 120 Skeleton
• NW-UCLA

NTU RGB+D 60 and 120

1. Request dataset here: https://rose1.ntu.edu.sg/dataset/actionRecognition
2. Download the skeleton-only datasets:
   1. nturgbd_skeletons_s001_to_s017.zip (NTU RGB+D 60)
   2. nturgbd_skeletons_s018_to_s032.zip (NTU RGB+D 120)
   3. Extract above files to ./data/nturgbd_raw

NW-UCLA

1. Download dataset from CTR-GCN
2. Move all_sqe to ./data/NW-UCLA

Data Processing

Directory Structure

Put downloaded data into the following directory structure:

- data/
  - NW-UCLA/
    - all_sqe
      ... # raw data of NW-UCLA
  - ntu/
  - ntu120/
  - nturgbd_raw/
    - nturgb+d_skeletons/     # from `nturgbd_skeletons_s001_to_s017.zip`
      ...
    - nturgb+d_skeletons120/  # from `nturgbd_skeletons_s018_to_s032.zip`
      ...

Generating Data

• Generate NTU RGB+D 60 or NTU RGB+D 120 dataset:

 cd ./data/ntu # or cd ./data/ntu120
 # Get skeleton of each performer
 python get_raw_skes_data.py
 # Remove the bad skeleton 
 python get_raw_denoised_data.py
 # Transform the skeleton to the center of the first frame
 python seq_transformation.py

Training & Testing

Training

bash train.sh

Please check the configuration in the config directory.

Testing

bash evaluate.sh

To ensemble the results of different modalities, run the following command:

bash ensemble.sh

Acknowledgements

This repo is based on 2s-AGCN and CTR-GCN. The data processing is borrowed from SGN and HCN, and the training strategy is based on Hyperformer.

Thanks to the original authors for their work!

References

@inproceedings{zhou2024blockgcn,
  title={BlockGCN: Redefining Topology Awareness for Skeleton-Based Action Recognition},
  author={Zhou, Yuxuan and Yan, Xudong and Cheng, Zhi-Qi and Yan, Yan and Dai, Qi and Hua, Xian-Sheng},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2024}
}

@article{zhou2023overcoming,
  title={Overcoming topology agnosticism: Enhancing skeleton-based action recognition through redefined skeletal topology awareness},
  author={Zhou, Yuxuan and Cheng, Zhi-Qi and He, Jun-Yan and Luo, Bin and Geng, Yifeng and Xie, Xuansong},
  journal={arXiv preprint arXiv:2305.11468},
  year={2023}
}