Department of Computer Engineering
S E M I N A R
Recognizing Human Actions Using Key Poses
Computer Engineering Department
In this study, we explore the idea of using only pose, without utilizing any temporal information, for human action recognition. In contrast to the other studies using complex action representations, we propose a simple method, which relies on extracting "key poses" from action sequences. Our contribution is two-fold. Firstly, representing the pose in a frame as a collection of line-pairs, we propose a matching scheme between two frames to compute their similarity. Secondly, after grouping the frames by k-medoids clustering to extract candidate key poses, we rank the potentiality of each candidate becoming a key pose, ensuring both their representative and discriminative properties. Our experimental results on KTH dataset have shown that pose information by itself is quite effective in grasping the nature of an action and sufficient to distinguish one from others.
DATE: 22 February, 2010, Monday @ 15:40
PLACE: EA 409