You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A contactless motion tracking pipeline built with Python and OpenCV. Utilizes MOG2 background segmentation, Farneback Optical Flow, and JET pseudocolour mapping for real-time motion intensity visualization and interactive gamification.
This contains code and dataset for the an enhanced spatio-temporal transformer framework for human motion quality assessment. The codes and results are available for our custom dataset and publicly available datasets (KIMORE and UI-PRMD).
High-performance ESP32 firmware for converting cross-trainer cadence into Bluetooth LE HID gamepad input. Features optimized axis mapping and NimBLE stability.