TY - GEN
T1 - Encoding spatio-temporal distribution by generalized VLAD for action recognition
AU - Sheng, Biyun
AU - Yan, Yan
AU - Sun, Changyin
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2015/6/19
Y1 - 2015/6/19
N2 - The location information of interest points is an important cue for action recognition. In order to model the spatio-temporal distribution, we propose a novel position feature which is constructed by normalized pairwise relative positions of points. Promising performance has been achieved by Vector of Locally Aggregated Descriptors (VLAD) which gather the differences between descriptors and visual words. However, original VLAD imposes equal weights for difference vectors and ignores zero-order statistics of local descriptors. In this paper, we present Generalized VLAD (GVLAD), an extension of VLAD to encode the position features as well as local appearance descriptors, by which different weights and zero-order information are simultaneously taken into consideration. The state-of-the-art performance on two benchmark datasets validates the effectiveness of our proposed method.
AB - The location information of interest points is an important cue for action recognition. In order to model the spatio-temporal distribution, we propose a novel position feature which is constructed by normalized pairwise relative positions of points. Promising performance has been achieved by Vector of Locally Aggregated Descriptors (VLAD) which gather the differences between descriptors and visual words. However, original VLAD imposes equal weights for difference vectors and ignores zero-order statistics of local descriptors. In this paper, we present Generalized VLAD (GVLAD), an extension of VLAD to encode the position features as well as local appearance descriptors, by which different weights and zero-order information are simultaneously taken into consideration. The state-of-the-art performance on two benchmark datasets validates the effectiveness of our proposed method.
UR - https://www.scopus.com/pages/publications/84938345121
U2 - 10.1109/CCECE.2015.7129346
DO - 10.1109/CCECE.2015.7129346
M3 - 会议稿件
AN - SCOPUS:84938345121
T3 - Canadian Conference on Electrical and Computer Engineering
SP - 620
EP - 625
BT - 2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering, CCECE 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2015 28th IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2015
Y2 - 3 May 2015 through 6 May 2015
ER -