https://github.com/davide710/pull-up_tracker
A custom trained object detection YOLO (v. 5) model that can recognize people doing pull-ups
https://github.com/davide710/pull-up_tracker
computer-vision object-detection object-tracking python yolo
Last synced: 2 months ago
JSON representation
A custom trained object detection YOLO (v. 5) model that can recognize people doing pull-ups
- Host: GitHub
- URL: https://github.com/davide710/pull-up_tracker
- Owner: davide710
- Created: 2024-02-28T17:20:21.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2025-01-16T17:24:11.000Z (over 1 year ago)
- Last Synced: 2025-03-23T19:15:10.054Z (over 1 year ago)
- Topics: computer-vision, object-detection, object-tracking, python, yolo
- Language: Jupyter Notebook
- Homepage:
- Size: 29.6 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# pull-up tracker
- a custom trained object detection YOLO (v. 5) model that can recognize people doing pull-ups with excellent results:
- a script (counting_single.py) that, given a video of one person training, can live count how many repetitions are being done
- a script (challenge.py) that tracks (with custom developed class Tracker()) how many pull-ups are executed by each person in the video, as in a competition
### Usage
```
git clone https://github.com/davide710/pull-up_tracker.git
cd pull-up_tracker
pip install numpy opencv-contrib-python pyyaml
# To execute a script with a video
python3 [script.py] /path/to/video
# To execute the script challenge.py with the example
python3 challenge.py videos/pullups_challenge.mp4
# To execute it with live webcam (you may need to adjust webcam configurations in the script)
python3 [script.py] -live
```
This is the result obtained with [videos/pullups_challenge.mp4](videos/pullups_challenge.mp4):
The original video was taken from https://www.youtube.com/shorts/ft7VmEyvcuc?feature=share
#