[ICCV2021] Official code for "Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition"

Last update: Dec 16, 2022

Related tags

Overview

CTR-GCN

This repo is the official implementation for Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition. The paper is accepted to ICCV2021.

Note: We also provide a simple and strong baseline model, which achieves 83.7% on NTU120 CSub with joint modality only, to facilitate the development of skeleton-based action recognition.

Architecture of CTR-GC

Prerequisites

Python >= 3.6
PyTorch >= 1.1.0
PyYAML, tqdm, tensorboardX
We provide the dependency file of our experimental environment, you can install all dependencies by creating a new anaconda virtual environment and running pip install -r requirements.txt
Run pip install -e torchlight

Data Preparation

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

Request dataset here: http://rose1.ntu.edu.sg/Datasets/actionRecognition.asp
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

Download dataset from here
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

Change the config file depending on what you want.

# Example: training CTRGCN on NTU RGB+D 120 cross subject with GPU 0
python main.py --config config/nturgbd120-cross-subject/default.yaml --work-dir work_dir/ntu120/csub/ctrgcn --device 0
# Example: training provided baseline on NTU RGB+D 120 cross subject
python main.py --config config/nturgbd120-cross-subject/default.yaml --model model.baseline.Model--work-dir work_dir/ntu120/csub/baseline --device 0

To train model on NTU RGB+D 60/120 with bone or motion modalities, setting bone or vel arguments in the config file default.yaml or in the command line.

# Example: training CTRGCN on NTU RGB+D 120 cross subject under bone modality
python main.py --config config/nturgbd120-cross-subject/default.yaml --train_feeder_args bone=True --test_feeder_args bone=True --work-dir work_dir/ntu120/csub/ctrgcn_bone --device 0

To train model on NW-UCLA with bone or motion modalities, you need to modify data_path in train_feeder_args and test_feeder_args to "bone" or "motion" or "bone motion", and run

python main.py --config config/ucla/default.yaml --work-dir work_dir/ucla/ctrgcn_xxx --device 0

To train your own model, put model file your_model.py under ./model and run:

# Example: training your own model on NTU RGB+D 120 cross subject
python main.py --config config/nturgbd120-cross-subject/default.yaml --model model.your_model.Model --work-dir work_dir/ntu120/csub/your_model --device 0

Testing

To test the trained models saved in <work_dir>, run the following command:

python main.py --config <work_dir>/config.yaml --work-dir <work_dir> --phase test --save-score True --weights <work_dir>/xxx.pt --device 0

To ensemble the results of different modalities, run

# Example: ensemble four modalities of CTRGCN on NTU RGB+D 120 cross subject
python ensemble.py --datasets ntu120/xsub --joint-dir work_dir/ntu120/csub/ctrgcn --bone-dir work_dir/ntu120/csub/ctrgcn_bone --joint-motion-dir work_dir/ntu120/csub/ctrgcn_motion --bone-motion-dir work_dir/ntu120/csub/ctrgcn_bone_motion

Pretrained Models

Download pretrained models for producing the final results on NTU RGB+D 60&120 cross subject [Google Drive].
Put files to <work_dir> and run Testing command to produce the final result.

Acknowledgements

This repo is based on 2s-AGCN. The data processing is borrowed from SGN and HCN.

Thanks to the original authors for their work!

Citation

Please cite this work if you find it useful:.

@article{chen2021channel,
  title={Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition},
  author={Chen, Yuxin and Zhang, Ziqi and Yuan, Chunfeng and Li, Bing and Deng, Ying and Hu, Weiming},
  journal={arXiv preprint arXiv:2107.12213},
  year={2021}
}

Contact

For any questions, feel free to contact: [email protected]

[ICCV2021] Official code for "Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition"

Related tags

Overview

CTR-GCN

Architecture of CTR-GC

Prerequisites

Data Preparation

Download datasets.

There are 3 datasets to download:

NTU RGB+D 60 and 120

NW-UCLA

Data Processing

Directory Structure

Generating Data

Training & Testing

Training

Testing

Pretrained Models

Acknowledgements

Citation

Contact

Owner

Yuxin Chen

Try out deep learning models online on Google Colab

PROJECT - Az Residential Real Estate Analysis

[NeurIPS 2021] The PyTorch implementation of paper "Self-Supervised Learning Disentangled Group Representation as Feature"

Official code for the ICLR 2021 paper Neural ODE Processes

SimDeblur is a simple framework for image and video deblurring, implemented by PyTorch

Yolo ros - YOLO-ROS for HUAWEI ATLAS200

Implementation of U-Net and SegNet for building segmentation

这是一个mobilenet-yolov4-lite的库，把yolov4主干网络修改成了mobilenet，修改了Panet的卷积组成，使参数量大幅度缩小。

Codebase for ECCV18 "The Sound of Pixels"

Implementation for paper LadderNet: Multi-path networks based on U-Net for medical image segmentation

Autonomous racing with the Anki Overdrive

Is RobustBench/AutoAttack a suitable Benchmark for Adversarial Robustness?

Easy Parallel Library (EPL) is a general and efficient deep learning framework for distributed model training.

The description of FMFCC-A (audio track of FMFCC) dataset and Challenge resluts.

Tesla Light Show xLights Guide With python

Codes for TIM2021 paper "Anchor-Based Spatio-Temporal Attention 3-D Convolutional Networks for Dynamic 3-D Point Cloud Sequences"

Seeing if I can put together an interactive version of 3b1b's Manim in Streamlit

Research using Cirq!

Resources related to our paper "CLIN-X: pre-trained language models and a study on cross-task transfer for concept extraction in the clinical domain"

Algorithmic encoding of protected characteristics and its implications on disparities across subgroups