This project is based on YOLOv3.
Please check the original YOLOv3 repository for better understanding.
detect.py
runs inference on any sources:
python3 detect.py --cfg cfg/yolov3-spp.cfg --weights weights/yolov3-spp-ultralytics.pt --source ...
- Image:
--source file.jpg
- Video:
--source file.mp4
- Directory:
--source dir/
- Webcam:
--source 0
- RTSP stream:
--source rtsp://170.93.143.139/rtplive/470011e600ef003a004ee33696235daa
- HTTP stream:
--source http://wmccpinetop.axiscam.net/mjpg/video.mjpg
You can download our pretrained weight from https://drive.google.com/file/d/1AR58mMSAZyAs0o_TMHBrdG9KSGQKQMZN/view?usp=sharing
download_video.py
download youtube video and mosaic video
- Video:
download_video.py "https://www.youtube.com/watch?v=YJZqi7LAW3E" "video"
For streaming videos, the download does not stop until the streaming is finished, so you need to set an end time.
- Streaming:
download_video.py "https://www.youtube.com/watch?v=YJZqi7LAW3E" "stream" 100
1. You need to change PROJECT_PATH in run_docker.sh
first.
SERVER=10.0
GPU=0
SHMEMORY=16g
PROJECT_PATH=/YOUR/PROJECT/PATH
2. Build docker
./build_docker.sh
3. Run docker
./run_docker.sh
...
...