Identification of Partitions in a Homogeneous Activity Group Using Mobile Devices

Abstract

People in public areas often appear in groups. People with homogeneous coarse-grained activities may be further divided into subgroups depending on more fine-grained behavioral differences. Automatically identifying these subgroups can benefit a variety of applications for group members. In this work, we focus on identifying such subgroups in a homogeneous activity group (i.e., a group of people who perform the same coarse-grained activity at the same time). We present a generic framework using sensors built in commodity mobile devices. Specifically, we propose a two-stage process, sensing modality selection given a coarse-grained activity, followed by multimodal clustering to identify subgroups. We develop one early fusion and one late fusion multimodal clustering algorithm. We evaluate our approaches using multiple datasets; two of them are with the same activity while the other has a different activity. The evaluation results show that the proposed multimodal-based approaches outperform existing work that uses only one single sensing modality and they also work in scenarios when manually selecting one sensing modality fails.

GARSAaaS: group activity recognition and situation analysis as a service

“…In terms of comparing with the state of the art, it is difficult to provide a fair quantitative comparison (such as classification accuracy or algorithm performance) given that there is no widely accepted benchmark in the area of sensor-based group activity recognition. Most of the studies simulated their experiments and executed their proposed algorithms offline, e.g., [7,15,27,35], in which, challenges such as the inaccuracy caused by communication delay has been disregarded. Also, having domaindependent parameters can influence results, such as window size variations.…”