Celloscope Exercise Monitoring System

Exercise feedback via vision-language models.

A system leveraging Vision-Language Models (VLMs) to assist users in performing exercises correctly by comparing their execution against reference videos of expert demonstrations. The system uses frame-level visual and motion comparison, integrated with language-based feedback, to generate natural language guidance that helps users improve their form and reduce the risk of injury.

Contributors

Abrar Ahmed1, Shahriar Kabir2

1Trainee-AI/ML, Celloscope 2Senior AI Research Engineer, Celloscope

What is Celloscope Exercise Monitoring System?

Celloscope Exercise Monitoring System is an intelligent system that helps users improve their exercise form by analyzing video input and providing personalized feedback on workout mistakes.

How It Works?

The system first extracts 2D keypoints from both the user and reference workout videos. These pose representations are then compared to detect alignment errors, independent of body type or appearance.

AI-Powered Feedback

We feed the pose comparison data into a vision-language model (VLM), which generates concise, human-readable feedback on form issues and how to fix them — like a smart personal trainer.

Video Analysis & Comparison

Reference Video - Text Removed

User Video - Text Removed

Reference Keypoints

User Keypoints