Real Time On Sensor Gait Phase Detection with 0.5KB Deep Learning Model
This enables efficient on-sensor processing for applications like wearable health monitoring, though it is incremental as it builds on existing U-Net architectures.
The paper tackled the problem of real-time gait phase detection on low-power sensors by developing a segmentation-based model that achieves 95.9% accuracy with only 0.5KB model size and 67K operations per second.
Gait phase detection with convolution neural network provides accurate classification but demands high computational cost, which inhibits real time low power on-sensor processing. This paper presents a segmentation based gait phase detection with a width and depth downscaled U-Net like model that only needs 0.5KB model size and 67K operations per second with 95.9% accuracy to be easily fitted into resource limited on sensor microcontroller.