Generative Resident Separation and Multi-label Classification for Multi-person Activity Recognition
This addresses activity recognition for multiple residents in smart homes, but it is incremental as it builds on existing methods and datasets.
The paper tackles multi-person activity recognition in smart homes by proposing Seq2Res for separating sensor events by resident and BiGRU+Q2L for multi-label classification, comparing them to a state-of-the-art model on a dataset of two residents.
This paper presents two models to address the problem of multi-person activity recognition using ambient sensors in a home. The first model, Seq2Res, uses a sequence generation approach to separate sensor events from different residents. The second model, BiGRU+Q2L, uses a Query2Label multi-label classifier to predict multiple activities simultaneously. Performances of these models are compared to a state-of-the-art model in different experimental scenarios, using a state-of-the-art dataset of two residents in a home instrumented with ambient sensors. These results lead to a discussion on the advantages and drawbacks of resident separation and multi-label classification for multi-person activity recognition.