Learning Multi-level Features For Sensor-based Human Action Recognition
This work addresses action recognition for applications like healthcare or fitness tracking, presenting a novel method for a known bottleneck in sensor-based analysis.
The paper tackles human action recognition from a single body-worn inertial sensor by proposing a multi-level feature learning framework, achieving state-of-the-art weighted F1 scores of 88.7%, 98.8%, and 72.6% on Skoda, WISDM, and OPP datasets.
This paper proposes a multi-level feature learning framework for human action recognition using a single body-worn inertial sensor. The framework consists of three phases, respectively designed to analyze signal-based (low-level), components (mid-level) and semantic (high-level) information. Low-level features capture the time and frequency domain property while mid-level representations learn the composition of the action. The Max-margin Latent Pattern Learning (MLPL) method is proposed to learn high-level semantic descriptions of latent action patterns as the output of our framework. The proposed method achieves the state-of-the-art performances, 88.7%, 98.8% and 72.6% (weighted F1 score) respectively, on Skoda, WISDM and OPP datasets.