TY - GEN
T1 - Hand posture recognition using hidden conditional random fields
AU - Liu, Te Cheng
AU - Wang, Ko Chih
AU - Tsai, Augustine
AU - Wang, Chieh Chih
PY - 2009
Y1 - 2009
N2 - Body-language understanding is essential to human robot interaction, and hand posture recognition is one of the most important components in a body-language recognition system. The existing hand posture recognition approaches based on robust local features such as SIFT can be invariant to background noise and in-plane rotation. However the ignorance of the relationships among local features is a fundamental issue. The part-based models argue that objects of the same category share the same part-structure which consists of parts and relationships among parts. In this paper, a discriminative partbased model, Hidden Conditional Random Fields (HCRFs), is used to recognize hand postures. Although the existing global locations of features have been used to consider large scale dependency among parts in the HCRFs framework, the results are not invariant to in-plane rotation. New features by the distance to the image center are proposed to encode the global relationship as well as to perform in-plane rotationinvariant recognition. The experimental results demonstrate that the proposed approach is in-plane rotation-invariant and out performs the approach using Ada Boost with SIFT.
AB - Body-language understanding is essential to human robot interaction, and hand posture recognition is one of the most important components in a body-language recognition system. The existing hand posture recognition approaches based on robust local features such as SIFT can be invariant to background noise and in-plane rotation. However the ignorance of the relationships among local features is a fundamental issue. The part-based models argue that objects of the same category share the same part-structure which consists of parts and relationships among parts. In this paper, a discriminative partbased model, Hidden Conditional Random Fields (HCRFs), is used to recognize hand postures. Although the existing global locations of features have been used to consider large scale dependency among parts in the HCRFs framework, the results are not invariant to in-plane rotation. New features by the distance to the image center are proposed to encode the global relationship as well as to perform in-plane rotationinvariant recognition. The experimental results demonstrate that the proposed approach is in-plane rotation-invariant and out performs the approach using Ada Boost with SIFT.
UR - http://www.scopus.com/inward/record.url?scp=70350462589&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70350462589&partnerID=8YFLogxK
U2 - 10.1109/AIM.2009.5229788
DO - 10.1109/AIM.2009.5229788
M3 - Conference contribution
AN - SCOPUS:70350462589
SN - 9781424428533
T3 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM
SP - 1828
EP - 1833
BT - 2009 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2009
T2 - 2009 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2009
Y2 - 14 July 2009 through 17 July 2009
ER -