Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[CVPR 2020 Oral] PyTorch implementation of "Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition"
| Date | Stars |
|---|---|
| 2026-07-30 | 458 |
| 2026-08-10 | 459 |
| 2026-09-03 | 460 |
| 2026-09-24 | 461 |
| 2026-09-28 | 462 |
| 2026-10-02 | 462 |
| 2026-10-28 | 461 |
| 2026-11-04 | 461 |
Today
— stars today
This week
— stars this week
This month
-1 stars this month
Momentum
0.0
growth rate 0.00%/day
# MS-G3D
PyTorch implementation of "Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition", CVPR 2020 Oral.
[[PDF](https://arxiv.org/pdf/2003.14111.pdf)][[Demo](https://youtu.be/5TcHIIece38)][[Abstract/Supp](https://openaccess.thecvf.com/content_CVPR_2020/html/Liu_Disentangling_and_Unifying_Graph_Convolutions_for_Skeleton-Based_Action_Recognition_CVPR_2020_paper.html)]
<img src="imgs/cropped-fixed-standup.gif" width="24%"><img src="imgs/cropped-fixed-clap.gif" width="24%">
<img src="imgs/cropped-fixed-salute.gif" width="24%">
<img src="imgs/cropped-fixed-pointfinger.gif" width="24%">
<img src="imgs/cropped-fixed-pickup.gif" width="24%">
<img src="imgs/cropped-fixed-rub.gif" width="24%">
<img src="imgs/cropped-fixed-drink.gif" width="24%">
<img src="imgs/cropped-fixed-checktime.gif" width="24%">
## Dependencies
- Python >= 3.6
- PyTorch >= 1.2.0
- [NVIDIA Apex](https://github.com/NVIDIA/apex) (auto mixed precision training)
- PyYAML, tqdm, tensorboardX
## Data Preparation
*Disk usage warning: after preprocessing, the total sizes of datasets are around 38GB, 77GB, 63GB for NTU RGB+D 60, NTU RGB+D 120, and Kinetics 400, respectively. The raw/intermediate sizes may be larger.*
### Download Datasets
There are 3 datasets to download:
- NTU RGB+D 60 Skeleton
- NTU RGB+D 120 Skeleton
- Kinetics 400 Skeleton
#### NTU RGB+D 60 and 120
1. Request dataset here: http://rose1.ntu.edu.sg/Datasets/actionRecognition.asp
2. Download the skeleton-only datasets:
- `nturgbd_skeletons_s001_to_s017.zip` (NTU RGB+D 60)
- `nturgbd_skeletons_s018_to_s032.zip` (NTU RGB+D 120, on top of NTU RGB+D 60)
- Total size should be 5.8GB + 4.5GB.
3. Download missing skeletons lookup files [from the authors' GitHub repo](https://github.com/shahroudy/NTURGB-D#samples-with-missing-skeletons):
- NTU RGB+D 60 Missing Skeletons:
`wget https://raw.githubusercontent.com/shahroudy/NTURGB-D/master/Matlab/NTU_RGBD_samples_with_missing_skeletons.txt`
- NTU RGB+D 120 Missing Skeletons:
`wget https://raw.githubusercontent.com/shahroudy/NTURGB-D/master/Matlab/NTU_RGBD120_samples_with_missing_skeletons.txt`
- Remember to remove the first few lines of text in these files!
#### Kinetics Skeleton 400
1. Download dataset from ST-GCN repo: https://github.com/yysijie/st-gcn/blob/master/OLD_README.md#kinetics-skeleton
2. [This](https://silicondales.com/tutorials/g-suite/how-to-wget-files-from-google-drive/) might be useful if you want to `wget` the dataset from Google Drive
### Data Preprocessing
#### Directory Structure
Put downloaded data into the following directory structure:
```
- data/
- kinetics_raw/
- kinetics_train/
...
- kinetics_val/
...
- kinetics_train_label.json
- keintics_val_label.json
- nturgbd_raw/
- nturgb+d_skeletons/ # from `nturgbd_skeletons_s001_to_s017.zip`
...
- nturgb+d_skeletons120/ # from `nturgbd_skeletons_s018_to_s032.zip`
...
- NTU_RGBD_samples_with_missing_skeletons.txt
- NTU_RGBD120_samples_with_missing_skeletons.txt
```
#### Generating Data
1. NTU RGB+D
- `cd data_gen`
- `python3 ntu_gendata.py`
- `python3 ntu120_gendata.py`
- Time estimate is ~ 3hrs to generate NTU 120 on a single core (feel free to parallelize the code :))
2. Kinetics
- `python3 kinetics_gendata.py`
- ~ 70 mins to generate Kinetics data
3. Generate the bone data with:
- `python gen_bone_data.py --dataset ntu`
- `python gen_bone_data.py --dataset ntu120`
- `python gen_bone_data.py --dataset kinetics`
## Pretrained Models
- Download pretrained models for producing the final results on NTU RGB+D 60, NTU RGB+D 120, Kinetics Skeleton 400: [[Dropbox](https://www.dropbox.com/s/9n9897cu1ft1khg/msg3d-pretrained-models.zip)][[GoogleDrive](https://drive.google.com/open?id=1y3VbEnINtyriy82apiTZJtBV1a3cywa-)][[WeiYun](https://share.weiyun.com/DlR4Jse1)]
- Put the folder of pretrained models at repo Excerpt of 8,225 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:1119cc729d51648b, topic:deep-learning, topic:pytorch
matched fp:1119cc729d51648b, topic:computer-vision