In this paper, we present a home-monitoring oriented human activityrecognition benchmark database, based on the combination of a colorvideo camera and a depth sensor. Our contributions are two-fold: 1) Wehave created a publicly releasable human activity video database (i.e.,named as RGBD-HuDaAct), which contains synchronized color-depth videostreams, for the task of human daily activity recognition. This databaseaims at encouraging more research efforts on human activity recognitionbased on multi-modality sensor combination (e.g., color plus depth). 2)Two multi-modality fusion schemes, which naturally combine color anddepth information, have been developed from two state-of-the-art featurerepresentation methods for action recognition, i.e., spatio-temporalinterest points (STIPs) and motion history images (MHIs). Thesedepth-extended feature representation methods are evaluatedcomprehensively and superior recognition performances over theiruni-modality (e.g., color only) counterparts are demonstrated.
CITATION STYLE
Ni, B., Wang, G., & Moulin, P. (2013). RGBD-HuDaAct: A Color-Depth Video Database for Human Daily Activity Recognition (pp. 193–208). https://doi.org/10.1007/978-1-4471-4640-7_10
Mendeley helps you to discover research relevant for your work.