Improved TZsearch Fast Search Algorithm for JMVC

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 163)


In this paper, we are concentrated on the research and improvement of the TZsearch fast search algorithm in JMVC. In JMVC, the same search algorithm was applied in the both inter-view and inter-frame. However, this paper applies different search algorithms on the reference frames of inter-view and the reference frames of inter-frame. On the premise of guaranteeing the quality of reconstructed video and coding bit-rate, the experiment result shows that in the test of sequences with different features the improved algorithm takes only 40–60 % total encoding time of the former fast algorithm.


Multi-view video coding Tzsearch Hybrid search algorithm 


  1. 1.
    Anthony Vetro, Yeping Su, Hideaki Kimata et al (2006) Joint Multi-view Video Model (JMVM) 1.0. JVT-T20S JulyGoogle Scholar
  2. 2.
    TUT/Thomson (2008) Joint multi-view video coding (JMVC) 1.0. May 31Google Scholar
  3. 3.
    Zhu Shah, Ma Kai-Kuang (2000) New diamond search algorithm for fast block-matching motion estimation. IEEE Trans Image Process 9(2):287–290CrossRefGoogle Scholar
  4. 4.
    Zhu, Ce, Lin, Xiao; Chau, Lap-Pui (2002) Hexagon-based search pattern for fast block motion estimation. IEEE Trans Circ Syst Video Tech 12(5):349–355CrossRefGoogle Scholar
  5. 5.
    Chen, Zhibo Xu, Jianfeng; He, Yun; Zheng, Junli (2006) Fast integer-pel and fractional-pel motion estimation for H.264/AVC. J Visual Commun Image Represent Emerg H.264/AVC Video Codin Stand 17(2):264–290Google Scholar
  6. 6.
    MERL has provided the ballroom, exit and vassar test sequences
  7. 7.
    Tseng Din-Chang, Chen yi-Lin, Liu MSC (2001) Wavelet-based multi-spectral image fusion. Inter Geosci Remote SensSymp 4:1956–1958Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  1. 1.Division of Information TechnologyGraduate School at Shenzhen, Tsinghua UniversityShenzhenChina

Personalised recommendations