DocumentCode :
2924411
Title :
Robust human pose estimation from corrupted images with partial occlusions and noise pollutions
Author :
Lu, Guoliang ; Kudo, Mineichi ; Toyama, Jun
Author_Institution :
Grad. Sch. of Inf. Sci. & Technol., Hokkaido Univ., Sapporo, Japan
fYear :
2011
fDate :
8-10 Nov. 2011
Firstpage :
433
Lastpage :
438
Abstract :
Robust human pose estimation from the given visual observations has attracted many attentions in the past two decades. However, this problem is still challenging due to the situation that observations are often corrupted with partial occlusions or noise pollutions or both in real-world applications. In this paper, we propose to estimate human pose by using robust silhouette matching in original rectangle-coordinate space. In addition, human action model is employed to determinate reasonable matching results. Experimental results on robustness sequence of Weizman dataset reveal that our proposed approach can estimate human pose robustly and reasonably when pose observations are corrupted with partial occlusions or noise pollutions.
Keywords :
gesture recognition; hidden feature removal; image matching; pose estimation; set theory; Weizman dataset sequence; corrupted images; human action model; noise pollutions; original rectangle-coordinate space; partial occlusions; robust human pose estimation; robust silhouette matching; Computational modeling; Estimation; Feature extraction; Humans; Legged locomotion; Robustness; Testing; affine transformation; corrupted visual observation; particle swarm optimization (PSO); pose estimation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Granular Computing (GrC), 2011 IEEE International Conference on
Conference_Location :
Kaohsiung
Print_ISBN :
978-1-4577-0372-0
Type :
conf
DOI :
10.1109/GRC.2011.6122636
Filename :
6122636
Link To Document :
بازگشت