TY - GEN
T1 - Asymmetry-based quality assessment of face images
AU - Zhang, Guangpeng
AU - Wang, Yunhong
PY - 2009
Y1 - 2009
N2 - Quality assessment plays an important role in biometrics field. Unlike the popularity of fingerprint and iris quality assessment, the evaluation of face quality is just started. To solve the incapability for performance prediction and remove the requirement for scale normalization of existing methods, three face quality measures are proposed in this paper. SIFT is utilized to extract scale insensitive feature points on face images, and three asymmetry-based quality measures are calculated by applying different constraints. Systematical experiments validate the efficacy of the proposed quality measures.
AB - Quality assessment plays an important role in biometrics field. Unlike the popularity of fingerprint and iris quality assessment, the evaluation of face quality is just started. To solve the incapability for performance prediction and remove the requirement for scale normalization of existing methods, three face quality measures are proposed in this paper. SIFT is utilized to extract scale insensitive feature points on face images, and three asymmetry-based quality measures are calculated by applying different constraints. Systematical experiments validate the efficacy of the proposed quality measures.
UR - https://www.scopus.com/pages/publications/72449211151
U2 - 10.1007/978-3-642-10520-3_47
DO - 10.1007/978-3-642-10520-3_47
M3 - 会议稿件
AN - SCOPUS:72449211151
SN - 364210519X
SN - 9783642105197
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 499
EP - 508
BT - Advances in Visual Computing - 5th International Symposium, ISVC 2009, Proceedings
T2 - 5th International Symposium on Advances in Visual Computing, ISVC 2009
Y2 - 30 November 2009 through 2 December 2009
ER -