P2LHAP:Wearable sensor-based human activity recognition, segmentation and forecast through Patch-to-Label Seq2Seq Transformer

arXiv:2403.08214v1 Announce Type: new
Abstract: Traditional deep learning methods struggle to simultaneously segment, recognize, and forecast human activities from sensor data. This limits their usefulness in many fields such as healthcare and assisted living, where real-time understanding of ongoing and upcoming activities is crucial. This paper introduces P2LHAP, a novel Patch-to-Label Seq2Seq framework that tackles all three tasks in a efficient single-task model. P2LHAP divides sensor data streams into a sequence of "patches", served as input tokens, and outputs a …

abstract arxiv cs.ai cs.cv data deep learning fields forecast healthcare human paper real-time recognition segment segmentation sensor seq2seq struggle through transformer type understanding wearable