TY - GEN
T1 - Fusing appearance features and correlation features for face video retrieval
AU - Jing, Chenchen
AU - Dong, Zhen
AU - Pei, Mingtao
AU - Jia, Yunde
N1 - Publisher Copyright:
© Springer International Publishing AG, part of Springer Nature 2018.
PY - 2018
Y1 - 2018
N2 - Face video retrieval has drawn considerable research attention recently. Most prior research mainly focused on either appearance features or correlation features, which could degrade retrieval performance. In this paper, we fuse appearance features and correlation features to exploit rich information of face videos for face video retrieval via a deep convolutional neural network. The network extracts appearance feature and correlation feature from a frame and the covariance matrix of a face video, respectively, and fuses them to obtain a comprehensive video representation. The fused feature is projected to a low-dimensional Hamming space via hash functions for the retrieval task. The network integrates feature extractions, feature fusion, and hash learning into a unified optimization framework to guarantee optimal compatibility of appearance features and correlation features. Experiments on two challenging TV-Series datasets demonstrate the effectiveness of the proposed method.
AB - Face video retrieval has drawn considerable research attention recently. Most prior research mainly focused on either appearance features or correlation features, which could degrade retrieval performance. In this paper, we fuse appearance features and correlation features to exploit rich information of face videos for face video retrieval via a deep convolutional neural network. The network extracts appearance feature and correlation feature from a frame and the covariance matrix of a face video, respectively, and fuses them to obtain a comprehensive video representation. The fused feature is projected to a low-dimensional Hamming space via hash functions for the retrieval task. The network integrates feature extractions, feature fusion, and hash learning into a unified optimization framework to guarantee optimal compatibility of appearance features and correlation features. Experiments on two challenging TV-Series datasets demonstrate the effectiveness of the proposed method.
KW - Appearance features
KW - Correlation features
KW - Deep CNN
KW - Face video retrieval
UR - http://www.scopus.com/inward/record.url?scp=85047444813&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-77383-4_15
DO - 10.1007/978-3-319-77383-4_15
M3 - Conference contribution
AN - SCOPUS:85047444813
SN - 9783319773827
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 150
EP - 160
BT - Advances in Multimedia Information Processing – PCM 2017 - 18th Pacific-Rim Conference on Multimedia, Revised Selected Papers
A2 - Zeng, Bing
A2 - Li, Hongliang
A2 - Huang, Qingming
A2 - El Saddik, Abdulmotaleb
A2 - Jiang, Shuqiang
A2 - Fan, Xiaopeng
PB - Springer Verlag
T2 - 18th Pacific-Rim Conference on Multimedia, PCM 2017
Y2 - 28 September 2017 through 29 September 2017
ER -