An open API service indexing awesome lists of open source software.

https://github.com/lukashedegaard/opendr-activity-recognition-demo

OpenDR human activity recognition demo
https://github.com/lukashedegaard/opendr-activity-recognition-demo

Last synced: 2 months ago
JSON representation

OpenDR human activity recognition demo

Awesome Lists containing this project

README

        

# OpenDR human activity recognition demo








Live demo of online human activity recognition using the [OpenDR toolkit](https://opendr.eu).

## Installation
```bash
git clone https://github.com/LukasHedegaard/opendr-activity-recognition-demo.git
cd opendr-activity-recognition-demo
pip install -e .
```

In addition, you need to install the OpenDR toolkit (_which is not publicly available yet_)

## Running the example
Human Activity Recognition using [X3D](https://openaccess.thecvf.com/content_CVPR_2020/papers/Feichtenhofer_X3D_Expanding_Architectures_for_Efficient_Video_Recognition_CVPR_2020_paper.pdf)
```bash
python demo.py --ip 0.0.0.0 --port 8000 --algorithm x3d --model xs
```

Human Activity Recognition using CoX3D
```bash
python demo.py --ip 0.0.0.0 --port 8000 --algorithm cox3d --model s
```

If you navigate your piano and http://0.0.0.0:8000 and pick up a ukulele, you might see something like this:

For other options, see `python demo.py --help`

## Troubleshooting
If no video is displayed for any of the methods, you may try to select another video source using the `--video_source` flag:
```bash
python demo.py --ip 0.0.0.0 --port 8000 --algorithm cox3d --model s --video_source 1
```

## Acknowledgement
This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871449 (OpenDR). This publication reflects the authors’ views only. The European Commission is not responsible for any use that may be made of the information it contains.