https://github.com/eugene87222/NCTU-Video-Streaming-and-Tracking
https://github.com/eugene87222/NCTU-Video-Streaming-and-Tracking
hls hls-live-streaming homework nctu object-tracking streaming yolov3
Last synced: 7 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/eugene87222/NCTU-Video-Streaming-and-Tracking
- Owner: eugene87222
- Created: 2020-12-30T09:11:06.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2021-01-14T10:08:26.000Z (over 4 years ago)
- Last Synced: 2024-08-06T23:37:27.525Z (10 months ago)
- Topics: hls, hls-live-streaming, homework, nctu, object-tracking, streaming, yolov3
- Language: Python
- Homepage:
- Size: 9.9 MB
- Stars: 4
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## Environment
- Ubuntu 18.04 or higher
- Python 3.7.5
## Requirements
- ffmpeg
- PyTorch 1.7+
- `pip install -r requirements.txt`
### Download YOLOv3 pretrained weight
```
mkdir yolo_weights
wget -P yolo_weights https://pjreddie.com/media/files/yolov3.weights
wget -P yolo_weights https://raw.githubusercontent.com/pjreddie/darknet/master/cfg/yolov3.cfg
wget -P yolo_weights https://raw.githubusercontent.com/pjreddie/darknet/master/data/coco.names
wget -P yolo_weights https://raw.githubusercontent.com/adipandas/multi-object-tracker/master/examples/pretrained_models/yolo_weights/coco_names.json
```
## Steps
1. Go to project directory
2. Execute `python http_server`
3. Execute `python server.py`
4. Execute `python stream.py`
5. Open `http://:/index.html` (`` and `http_port` are defined in `config.json`).
## Reference
- UI design:
- https://freshman.tech/custom-html5-video/
- https://www.w3schools.com/css/css_dropdowns.asp
- HLS streaming: https://video.aminyazdanpanah.com/python/start
- Flask server: https://towardsdatascience.com/video-streaming-in-web-browsers-with-opencv-flask-93a38846fe00