REFIT fixes wearable sensor placement errors without labels or retraining
REFIT: Recognize, Fix, and Test Wearable Sensor Placement Shifts without Labels
Machine LearningHuman-Computer Interaction
Summary
Wearable activity trackers like smartwatches often rely on sensors to recognize what you are doing. However, if the sensor is worn differently than usual—such as switching wrists or flipping the strap—the model might get confused because the sensor axes have changed. The authors present REFIT, a method that automatically detects these shifts and corrects the input data without needing any new labels or retraining. It then tests whether the correction worked and asks users to adjust the sensor if needed, helping keep activity recognition accurate.
What this means in practice
- •For wearable device developers: Maintain high activity recognition accuracy despite variations in how users wear devices by automatically correcting sensor data during use.
- •For fitness app developers: Reduce dependency on recalibrating or retraining activity models when users switch or adjust wearable sensors, improving user experience without extra labeling.
Authors
Bangxun Tang
Abstract
We present REFIT, an input calibration for frozen activity-recognition models whose inertial sensors are worn differently at deployment than in training. When users move a watch to the other wrist or put a strap sensor back on turned, the model sees the same motion on changed axes. REFIT undoes such shifts without labels or retraining. It describes them by families of axis transforms, such as reflections and rotations, and fits each family to the user's data so that simple statistics match those of the training data. The family that removes most of the mismatch names the shift. REFIT fixes the shift by applying the best member of that family before the frozen model and re-estimating its normalization statistics. It tests the fixed model with a label-free accuracy estimate and asks the user to re-wear the sensor when it is low. Experiments on real left/right sensor pairs and on real and simulated re-attachment show that REFIT outperforms label-free test-time adaptation methods on every dataset and restores most of the accuracy lost to re-attachment. It names injected shifts far more reliably than a confidence-based selector. After a correction over all signed permutations of the axes, the estimate separates successful from failed corrections.