WearableBridge

Code and interactive figures for WearableBridge: Task-Guided Latent Transport for Free-Living Activity Recognition.

Thigh2Wrist

Wrist features before and after MeanFlow adaptation, with concurrent measured-thigh features.

Open Thigh2Wrist

Lab2Freeliving

Wrist features before and after Mapper, MeanFlow, and Distillation, with independent laboratory references.

Open Lab2Freeliving

Code

Prepare features and a fitted reference classifier, then train a bridge:

python -m pip install -r requirements.txt
python train.py --data calibration.pt --output bridge.pt --method meanflow

Instructions · Bridge models · Training script