rTsfNet: a DNN model with Multi-head 3D Rotation and Time Series Feature Extraction for IMU-based Human Activity Recognition
This work addresses a domain-specific issue in human activity recognition by improving accuracy for applications like healthcare or fitness tracking, though it is incremental as it builds on existing deep learning and feature extraction methods.
The paper tackled the problem of inconsistent feature extraction in IMU-based human activity recognition by proposing rTsfNet, a DNN model that automatically selects 3D bases and integrates time series features, achieving higher accuracy than existing models on multiple benchmark datasets including UCI HAR, PAMAP2, Daphnet, and OPPORTUNITY.
Although many deep learning (DL) algorithms have been proposed for the IMU-based HAR domain, traditional machine learning that utilizes handcrafted time series features (TSFs) still often performs well. It is not rare that combinations among DL and TSFs show better accuracy than DL-only approaches. However, there is a problem with time series features in IMU-based HAR. The amount of derived features can vary greatly depending on the method used to select the 3D basis. Fortunately, DL's strengths include capturing the features of input data and adaptively deriving parameters. Thus, as a new DNN model for IMU-based human activity recognition (HAR), this paper proposes rTsfNet, a DNN model with Multi-head 3D Rotation and Time Series Feature Extraction. rTsfNet automatically selects 3D bases from which features should be derived by extracting 3D rotation parameters within the DNN. Then, time series features (TSFs), based on many researchers' wisdom, are derived to achieve HAR using MLP. Although rTsfNet is a model that does not use CNN, it achieved higher accuracy than existing models under well-managed benchmark conditions and multiple datasets: UCI HAR, PAMAP2, Daphnet, and OPPORTUNITY, all of which target different activities.