Paper Info
Title
Monocular Human Action Recognition Utilizing Silhouette Feature Extraction and Skin Color Detection
Abstract
Exemplar-based methods have been widely used in human action recognition. To analyze human action in monocular video has always been a challenging problem, due to depth information loss and ambiguities. In this paper we presented a method applying skin color detection and then calculating relative positions of face and hands to solve self-occlusions and to eliminate ambiguities. Then we applied 2D shape analysis to classify basic human actions. Several low level features were used to describe shapes, which needs less computation and can improve recognition speed to real-time level. We testified our method on a public action database and got satisfying results.
Year
DOI
Venue
2012
10.1109/PDCAT.2012.98
PDCAT
Keywords
Field
DocType
silhouette feature,self-occlusions,silhouette feature extraction,relative face positions,human action classification,2d shape analysis,recognition speed,basic human action,public action database,relative hands positions,real-time level,exemplar-based methods,depth information loss,key frame extraction,human action recognition,feature extraction,image classification,challenging problem,skin color detection,face and hands detection,human action analysis,ambiguity elimination,low-level features,utilizing silhouette feature extraction,gesture recognition,monocular human action recognition,human action,low level feature,exemplar-based method,skin,image colour analysis,monocular video
Computer vision,Three-dimensional face recognition,Pattern recognition,Silhouette,Computer science,Feature (computer vision),Gesture recognition,Feature extraction,Artificial intelligence,Contextual image classification,Monocular,Shape analysis (digital geometry)
Conference
ISBN
Citations 
PageRank 
978-0-7695-4879-1
1
0.38
References 
Authors
8
4
Name
Order
Citations
PageRank
Junjie Zhang110.38
Rentao Gu2258.24
Qing Ye310.71
Yuefeng Ji430349.02