reevald/ai-workout-assistant
AI-based pose tracking and repetitions counter to help everyone do the workout.
Leverages TensorFlow.js MoveNet for client-side pose detection that extracts keypoints from video/webcam input, then feeds them into a custom Dense Neural Network classifier to identify workout types and count repetitions. Supports custom workout training through a Google Colab pipeline where users can collect keypoint datasets, train binary classification models, and configure stage-based rep counting via angle thresholds on specific joint pairs.
226 stars. No commits in the last 6 months.
Stars
226
Forks
42
Language
JavaScript
License
MIT
Category
Last pushed
May 25, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/reevald/ai-workout-assistant"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
RiccardoRiccio/Fitness-AI-Trainer-With-Automatic-Exercise-Recognition-and-Counting
An extension of the previous 'Fitness-AI-Coach': a complete web application with real-time...
NgoQuocBao1010/Exercise-Correction
Make use of the power of Mediapipe’s pose detection, this project is built in order to analyze,...
cris-maillo/yogAI
work in progress
Furkan-Gulsen/Sport-With-AI
The human body is detected with the help of the Mediapipe library. Then, using the mathematical...
chrisprasanna/Exercise_Recognition_AI
Developing a virtual personal fitness tracker and exercise activity recognition A.I. using...