Project · 2024–25
← All workReviva
Post-stroke rehabilitation from a smartphone camera. OpenCV movement recognition plus a custom PyTorch model that gives patients real-time exercise feedback from visual data alone.
Winner, Sopra Steria International Student Challenge
- Python
- OpenCV
- PyTorch
- Hugging Face
- scikit-learn
Overview
Reviva turns a smartphone camera into a post-stroke rehab coach. Computer vision tracks each movement and gives the patient feedback in real time.
The problem
- Real-time movement recognition from smartphone video input
- Ensuring accuracy in detecting subtle rehabilitation movements
- Coordinating data preparation, model training and app integration across subteams
What we built
- Co-led the ML and computer-vision pipeline for patient movement recognition from smartphone camera data (OpenCV, PyTorch)
- Trained a model that gives real-time exercise feedback from visual input only
Results
- Winner, Sopra Steria International Student Challenge
- 3rd of 850 at EIT Health i-Days 2025
- Real-time exercise feedback from smartphone camera data alone