Hash Bit Selection for Nearest Neighbor Search

被引:36
|
作者
Liu, Xianglong [1 ]
He, Junfeng [2 ]
Chang, Shih-Fu [3 ]
机构
[1] Beihang Univ, State Key Lab Software Dev Environm, Beijing 10091, Peoples R China
[2] Google, Mountain View, CA 94043 USA
[3] Columbia Univ, Dept Elect Engn, New York, NY 10027 USA
基金
中国国家自然科学基金;
关键词
Nearest neighbor search; hash bit selection; bit reliability; bit complementarity; normalized dominant set; locality-sensitive hashing; QUANTIZATION;
D O I
10.1109/TIP.2017.2695895
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
To overcome the barrier of storage and computation when dealing with gigantic-scale data sets, compact hashing has been studied extensively to approximate the nearest neighbor search. Despite the recent advances, critical design issues remain open in how to select the right features, hashing algorithms, and/or parameter settings. In this paper, we address these by posing an optimal hash bit selection problem, in which an optimal subset of hash bits are selected from a pool of candidate bits generated by different features, algorithms, or parameters. Inspired by the optimization criteria used in existing hashing algorithms, we adopt the bit reliability and their complementarity as the selection criteria that can be carefully tailored for hashing performance in different tasks. Then, the bit selection solution is discovered by finding the best tradeoff between search accuracy and time using a modified dynamic programming method. To further reduce the computational complexity, we employ the pairwise relationship among hash bits to approximate the highorder independence property, and formulate it as an efficient quadratic programming method that is theoretically equivalent to the normalized dominant set problem in a vertex-and edge-weighted graph. Extensive large-scale experiments have been conducted under several important application scenarios of hash techniques, where our bit selection framework can achieve superior performance over both the naive selection methods and the state-of-the-art hashing algorithms, with significant accuracy gains ranging from 10% to 50%, relatively.
引用
收藏
页码:5367 / 5380
页数:14
相关论文
共 50 条
  • [1] Bit selection via walks on graph for hash-based nearest neighbor search
    Wang, Ya
    Liu, Xianglong
    Zhang, Danchen
    Wang, Deqing
    Yang, Yuan
    NEUROCOMPUTING, 2016, 213 : 137 - 146
  • [2] Query-Adaptive Reciprocal Hash Tables for Nearest Neighbor Search
    Liu, Xianglong
    Deng, Cheng
    Lang, Bo
    Tao, Dacheng
    Li, Xuelong
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2016, 25 (02) : 907 - 919
  • [3] Online learning to Hash for nearest neighbor search
    Qian J.-B.
    Hu W.
    Chen H.-H.
    Dong Y.-H.
    Kongzhi yu Juece/Control and Decision, 2019, 34 (12): : 2567 - 2575
  • [4] Boosting Complementary Hash Tables for Fast Nearest Neighbor Search
    Liu, Xianglong
    Deng, Cheng
    Mu, Yadong
    Li, Zhujin
    THIRTY-FIRST AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2017, : 4183 - 4189
  • [5] Multi-View Complementary Hash Tables for Nearest Neighbor Search
    Liu, Xianglong
    Huang, Lei
    Deng, Cheng
    Lu, Jiwen
    Lang, Bo
    2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2015, : 1107 - 1115
  • [6] DOUBLE-BIT QUANTIZATION AND WEIGHTING FOR NEAREST NEIGHBOR SEARCH
    Deng, Han
    Xie, Hongtao
    Ma, Wei
    Mao, Zhendong
    Zhou, Chuan
    2017 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2017, : 1717 - 1721
  • [7] Query-Adaptive Hash Code Ranking for Fast Nearest Neighbor Search
    Ji, Tianxu
    Liu, Xianglong
    Deng, Cheng
    Huang, Lei
    Lang, Bo
    PROCEEDINGS OF THE 2014 ACM CONFERENCE ON MULTIMEDIA (MM'14), 2014, : 1005 - 1008
  • [8] Nearest Neighbor Search Based on Bit String Partition and Multiple Index
    Miao J.
    Li Z.
    Zhou Z.
    Yang C.
    Liu Z.
    Liu W.
    Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao/Journal of Computer-Aided Design and Computer Graphics, 2019, 31 (05): : 771 - 779
  • [9] Double-Bit Quantization and Index Hashing for Nearest Neighbor Search
    Xie, Hongtao
    Mao, Zhendong
    Zhang, Yongdong
    Deng, Han
    Yan, Chenggang
    Chen, Zhineng
    IEEE TRANSACTIONS ON MULTIMEDIA, 2019, 21 (05) : 1248 - 1260
  • [10] General Distributed Hash Learning on Image Descriptors for k-Nearest Neighbor Search
    Cao, Yuan
    Qi, Heng
    Gui, Jie
    Li, Shuai
    Li, Keqiu
    IEEE SIGNAL PROCESSING LETTERS, 2019, 26 (05) : 750 - 754