Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/huguyuehuhu/HCN-pytorch
A pytorch reproduction of { Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation }.
https://github.com/huguyuehuhu/HCN-pytorch
hcn nturgb-d pytorch-implmention skeleton-based-action-recognition
Last synced: 2 months ago
JSON representation
A pytorch reproduction of { Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation }.
- Host: GitHub
- URL: https://github.com/huguyuehuhu/HCN-pytorch
- Owner: huguyuehuhu
- Created: 2018-08-27T14:05:13.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2022-11-22T02:35:11.000Z (over 1 year ago)
- Last Synced: 2024-01-26T20:34:50.646Z (5 months ago)
- Topics: hcn, nturgb-d, pytorch-implmention, skeleton-based-action-recognition
- Language: Python
- Homepage:
- Size: 355 KB
- Stars: 217
- Watchers: 7
- Forks: 61
- Open Issues: 14
-
Metadata Files:
- Readme: README.md
Lists
- Awesome-pytorch-list - HCN-pytorch - occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation }. (Paper implementations / Other libraries:)
- Awesome-pytorch-list-CNVersion - HCN-pytorch - occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation }. (Paper implementations｜论文实现 / Other libraries｜其他库:)
README
# A PyTorch Reproduction of HCN
**Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation**.
Chao Li, Qiaoyong Zhong, Di Xie, Shiliang Pu, IJCAI 2018.
[Arxiv Preprint](http://arxiv.org/pdf/1804.06055.pdf)
## Features
#### 1. Dataset
- [x] NTU RGB+D: Cross View (CV), Cross Subject (CS)
- [ ] SBU Kinect Interaction
- [ ] PKU-MMD
#### 2. Tasks
- [x] Action recognition
- [ ] Action detection
#### 3. Visualization
- Visdom supported.
## Prerequisites
Our code is based on **Python3.5**. There are a few dependencies to run the code in the following:
- Python >= 3.5
- **PyTorch == 0.4.0**
- [torchnet](https://github.com/pytorch/tnt)
- Visdom
- Other version info about some Python packages can be found in `requirements.txt`
## Usage
#### Data preparation
##### NTU RGB+D
To transform raw NTU RGB+D data into numpy array (memmap format ) by this command:
```commandline
python ./feeder/ntu_gendata.py --data_path --out_folder
```
##### Other Datasets
Not supported now.
#### Training
Before you start the training, you have to launch [visdom](https://github.com/facebookresearch/visdom) server.
```commandline
python -m visdom
```
To train the model, you should note that:
- ```--dataset_dir``` is the **parents path** for **all** the datasets,
- ``` --num ``` the number of experiments trials (type: list).
```commandline
python main.py --dataset_dir --mode train --model_name HCN --dataset_name NTU-RGB-D-CV --num 01
```
To run a new trial with different parameters, you need to:
- Firstly, run the above training command with a new trial number, e.g, ```--num 03```, thus you will got an error.
- Secondly, copy a parameters file from the ```./HCN/experiments/NTU-RGB-D-CV/HCN01/params.json``` to the path of your new trial ```"./HCN/experiments/NTU-RGB-D-CV/HCN03/params.json"``` and modify it as you want.
- At last, run the above training command again, it will works.
#### Testing
```commandline
python main.py --dataset_dir --mode test --load True --model_name HCN --dataset_name NTU-RGB-D-CV --num 01
```
#### Load and Training
You also can load a half trained model, and start training it from a specific checkpoint by the following command:
```commandline
python main.py --dataset_dir --mode load_train --load True --model_name HCN --dataset_name NTU-RGB-D-CV --num 01 --load_model
```
## Results
#### Table
The expected **Top-1** **accuracy** of the model for NTU-RGD+D are shown here (There is an **accuracy gap**. I am not the author of original HCN paper, the repo was reproduced according to the paper text and have not been tuned carefully):
| Model | Normalized
Sequence
Length | FC
Neuron
Numbers | NTU RGB+D
Cross Subject (%) |NTU RGB+D
Cross View (%) |
| :------: | :------: | :------:| :------:| :------: |
| HCN[1]| 32 | 256 | **86.5** | **91.1** |
| HCN | 32 | 256 | 84.2 | 89.2 |
| HCN | 64 | 512 | 84.9* | 90.9* |
[1] http://arxiv.org/pdf/1804.06055.pdf
#### Figures
- Loss & accuracy[CV]
#### Confusion matrix
- Loss & accuracy[CS]
## Reference
[1] Chao Li, Qiaoyong Zhong, Di Xie, Shiliang Pu. Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation. IJCAI 2018.
[2] [yysijie/st-gcn](https://github.com/yysijie/st-gcn): referred for some code of dataset processing.