Generalization of Fitness Exercise Recognition from Doppler Measurements by Domain-adaption and Few-Shot Learning

TL;DR

This paper addresses the challenge of generalizing fitness exercise recognition from Doppler measurements across diverse environments and users by employing domain adaptation and few-shot learning techniques, significantly improving accuracy.

Contribution

It introduces a new database with controlled and uncontrolled exercise data and proposes methods to adapt models with minimal data, enhancing real-world applicability.

Findings

Recognition accuracy increased by 2 to 6 times with adaptation methods

Proposed approaches effectively handle user, environment, and device variations

Database enables studying generalization in diverse scenarios

Abstract

In previous works, a mobile application was developed using an unmodified commercial off-the-shelf smartphone to recognize whole-body exercises. The working principle was based on the ultrasound Doppler sensing with the device built-in hardware. Applying such a lab-environment trained model on realistic application variations causes a significant drop in performance, and thus decimate its applicability. The reason of the reduced performance can be manifold. It could be induced by the user, environment, and device variations in realistic scenarios. Such scenarios are often more complex and diverse, which can be challenging to anticipate in the initial training data. To study and overcome this issue, this paper presents a database with controlled and uncontrolled subsets of fitness exercises. We propose two concepts to utilize small adaption data to successfully improve model generalization in an uncontrolled environment, increasing the recognition accuracy by two to six folds compared to the baseline for different users.