Exploiting spatial-temporal context for trajectory based action video retrieval

被引:4
作者
Zhang, Lelin [1 ]
Wang, Zhiyong [1 ]
Yao, Tingting [1 ,2 ]
Staoh, Shin'ichi [3 ]
Mei, Tao [4 ]
Feng, David Dagan [1 ]
机构
[1] Univ Sydney, Sch Informat Technol, Sydney, NSW, Australia
[2] Hefei Univ Technol, Sch Comp & Informat, Hefei, Anhui, Peoples R China
[3] Natl Inst Informat, Tokyo, Japan
[4] Microsoft Res, Beijing, Peoples R China
基金
澳大利亚研究理事会;
关键词
Spatial-temporal information; Descriptor coding; Trajectory matching; Bag-of-visual-words; Action video retrieval; EXTRACTION; SELECTION; MODEL;
D O I
10.1007/s11042-017-4353-2
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Retrieving videos with similar actions is an important task with many applications. Yet it is very challenging due to large variations across different videos. While the state-of-the-art approaches generally utilize the bag-of-visual-words representation with the dense trajectory feature, the spatial-temporal context among trajectories is overlooked. In this paper, we propose to incorporate such information into the descriptor coding and trajectory matching stages of the retrieval pipeline. Specifically, to capture the spatial-temporal correlations among trajectories, we develop a descriptor coding method based on the correlation between spatial-temporal and feature aspects of individual trajectories. To deal with the mis-alignments between dense trajectory segments, we develop an offset-aware distance measure for improved trajectory matching. Our comprehensive experimental results on two popular datasets indicate that the proposed method improves the performance of action video retrieval, especially on more dynamic actions with significant movements and cluttered backgrounds.
引用
收藏
页码:2057 / 2081
页数:25
相关论文
共 56 条
  • [1] [Anonymous], CORR
  • [2] [Anonymous], 2010, LECT NOTES COMPUT SC
  • [3] [Anonymous], P IEEE 7 WORKSH MULT
  • [4] [Anonymous], 2015, IEEE 17 INT WORKSH M
  • [5] [Anonymous], INT JOINT C ART INT
  • [6] [Anonymous], ACM INT C MULT RETR
  • [7] [Anonymous], IEEE T KNOWLEDGE DAT
  • [8] Real-time motion trajectory-based indexing and retrieval of video sequences
    Bashir, Faisal I.
    Khokhar, Ashfaq A.
    Schonfeld, Dan
    [J]. IEEE TRANSACTIONS ON MULTIMEDIA, 2007, 9 (01) : 58 - 65
  • [9] Bregonzio M, 2009, PROC CVPR IEEE, P1948, DOI 10.1109/CVPRW.2009.5206779
  • [10] Extraction and clustering of motion trajectories in video
    Buzan, D
    Sclaroff, S
    Kollios, G
    [J]. PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 2, 2004, : 521 - 524