TY - GEN
T1 - Underdetermined approach to real-time face tracking and recognition
AU - Chugan, Hisayoshi
AU - Oka, Yuki
AU - Shakunaga, Takeshi
N1 - Funding Information:
This work has been supported in part by a Grant-In-Aid for Scientific Research (No.20300067) from the Ministry of Education, Science, Sports, and Culture of Japan.
Publisher Copyright:
© 2013; MVA Organization. All rights reserved.
PY - 2013
Y1 - 2013
N2 - In a real-time face tracking and recognition system proposed by Oka and Shakunaga, an optimum weighted average of registered images are estimated and the weights are used for face identification and shape inference.Although their method works well even when a target face changes pose in photometric changes, both the person identification and expression recognition could not be robustly solved at the same time because of a capacity problem. This paper proposes underdetermined approach to solve the capacity problem. Although a single underdetermined system often results in some performance reduction, parallel implementation can remarkably improve the performance. Experimental results showed that the proposed method successfully worked for 10-person discrimination when 10 expressions were registered for each person even when an image sequence included many face motion, expression, photometric change and occasional occlusions.
AB - In a real-time face tracking and recognition system proposed by Oka and Shakunaga, an optimum weighted average of registered images are estimated and the weights are used for face identification and shape inference.Although their method works well even when a target face changes pose in photometric changes, both the person identification and expression recognition could not be robustly solved at the same time because of a capacity problem. This paper proposes underdetermined approach to solve the capacity problem. Although a single underdetermined system often results in some performance reduction, parallel implementation can remarkably improve the performance. Experimental results showed that the proposed method successfully worked for 10-person discrimination when 10 expressions were registered for each person even when an image sequence included many face motion, expression, photometric change and occasional occlusions.
UR - http://www.scopus.com/inward/record.url?scp=84969374474&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84969374474&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84969374474
SN - 9784901122139
T3 - Proceedings of the 13th IAPR International Conference on Machine Vision Applications, MVA 2013
SP - 242
EP - 246
BT - Proceedings of the 13th IAPR International Conference on Machine Vision Applications, MVA 2013
PB - MVA Organization
T2 - 13th IAPR International Conference on Machine Vision Applications, MVA 2013
Y2 - 20 May 2013 through 23 May 2013
ER -