Beyond Isolated Frames: Enhancing Sensor-Based Human Activity Recognition through Intra- and Inter-Frame Attention
This work addresses the need for better temporal modeling in HAR for applications like healthcare and sports, but it appears incremental as it builds on existing ConvNet approaches.
The paper tackles the problem of sensor-based Human Activity Recognition (HAR) by addressing the limitation of frame-by-frame analysis in Convolutional Neural Networks, which overlooks temporal dynamics, and proposes an intra- and inter-frame attention model with a time-sequential batch learning strategy to enhance recognition.
Human Activity Recognition (HAR) has become increasingly popular with ubiquitous computing, driven by the popularity of wearable sensors in fields like healthcare and sports. While Convolutional Neural Networks (ConvNets) have significantly contributed to HAR, they often adopt a frame-by-frame analysis, concentrating on individual frames and potentially overlooking the broader temporal dynamics inherent in human activities. To address this, we propose the intra- and inter-frame attention model. This model captures both the nuances within individual frames and the broader contextual relationships across multiple frames, offering a comprehensive perspective on sequential data. We further enrich the temporal understanding by proposing a novel time-sequential batch learning strategy. This learning strategy preserves the chronological sequence of time-series data within each batch, ensuring the continuity and integrity of temporal patterns in sensor-based HAR.