Human-Object Interaction Recognition by Learning the distances between the Object and the Skeleton Joints

Meng Meng (1) , Hassen Drira (2) , Mohamed Daoudi (2) , Jacques Boonaert (3)

1 TUST - Tianjin University of Science and Technology
2 CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189
3 CERI SN - IMT Nord Europe - Centre for Digital Systems

— In this paper we present a fully automatic approach for human-object interaction recognition from depth sensors. Towards that goal, we extract relevant frame-level features such as inter-joint distances and joint-object distances that are suitable for real time action recognition. These features are insensitive to position and pose variation. Experiments conducted on ORGBD dataset following state-of-the-art settings show the effectiveness of the proposed approach.

Domaines: Informatique [cs]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Meng Meng, Hassen Drira, Mohamed Daoudi, Jacques Boonaert. Human-Object Interaction Recognition by Learning the distances between the Object and the Skeleton Joints. Face and Gesture, 2015, Ljubljana, Slovenia. ⟨hal-01703222⟩