Learning to assign binary weights to binary descriptor

被引:0
|
作者
Huang, Zhoudi [1 ,2 ]
Wei, Zhenzhong [1 ,2 ]
Zhang, Guangjun [1 ,2 ]
机构
[1] Beihang Univ, Sch Instrumentat Sci & Optoelect Engn, Beijing 100191, Peoples R China
[2] Beihang Univ, Minist Educ, Key Lab Precis Optomechatron Technol, Beijing 100191, Peoples R China
来源
INFRARED TECHNOLOGY AND APPLICATIONS, AND ROBOT SENSING AND ADVANCED CONTROL | 2016年 / 10157卷
关键词
Binary local feature descriptor; binary weight; large-scale regularized optimization; binary approximation; fast matching;
D O I
10.1117/12.2246737
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Constructing robust binary local feature descriptors are receiving increasing interest due to their binary nature, which can enable fast processing while requiring significantly less memory than their floating-point competitors. To bridge the performance gap between the binary and floating-point descriptors without increasing the computational cost of computing and matching, optimal binary weights are learning to assign to binary descriptor for considering each bit might contribute differently to the distinctiveness and robustness. Technically, a large-scale regularized optimization method is applied to learn float weights for each bit of the binary descriptor. Furthermore, binary approximation for the float weights is performed by utilizing an efficient alternatively greedy strategy, which can significantly improve the discriminative power while preserve fast matching advantage. Extensive experimental results on two challenging datasets(Brown dataset and Oxford dataset) demonstrate the effectiveness and efficiency of the proposed method.
引用
收藏
页数:8
相关论文
共 50 条
  • [31] ON THE CAPACITY OF NEURAL NETWORKS WITH BINARY WEIGHTS
    KOCHER, I
    MONASSON, R
    JOURNAL OF PHYSICS A-MATHEMATICAL AND GENERAL, 1992, 25 (02): : 367 - 380
  • [32] Binary Linear Codes With Three Weights
    Ding, Kelan
    Ding, Cunsheng
    IEEE COMMUNICATIONS LETTERS, 2014, 18 (11) : 1879 - 1882
  • [33] Binary Linear Codes With Two Weights
    Wang, Qiuyan
    Ding, Kelan
    Xue, Rui
    IEEE COMMUNICATIONS LETTERS, 2015, 19 (07) : 1097 - 1100
  • [34] Automatic porting of binary file descriptor library
    Abbaspour, M
    Zhu, JW
    SYSTEM-ON-CHIP FOR REAL-TIME APPLICATIONS, 2003, : 193 - 202
  • [35] Color Binary Correlation descriptor for Image Retrieval
    Wu, Jun
    Liu, Shenglan
    Feng, Lin
    PROCEEDINGS OF THE 2016 12TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION (WCICA), 2016, : 1837 - 1841
  • [36] DIRECTED DRIFT - A NEW LINEAR THRESHOLD ALGORITHM FOR LEARNING BINARY WEIGHTS ONLINE
    VENKATESH, SS
    JOURNAL OF COMPUTER AND SYSTEM SCIENCES, 1993, 46 (02) : 198 - 217
  • [37] BISNN: TRAINING SPIKING NEURAL NETWORKS WITH BINARY WEIGHTS VIA BAYESIAN LEARNING
    Jang, Hyeryung
    Skatchkovsky, Nicolas
    Simeone, Osvaldo
    2021 IEEE DATA SCIENCE AND LEARNING WORKSHOP (DSLW), 2021,
  • [38] CS-FREAK: An improved binary descriptor
    Wang, Jianyong, 1600, Springer Verlag (437):
  • [39] Discriminative Binary Descriptor for Finger Vein Recognition
    Liu, Haiying
    Yang, Lu
    Yang, Gongping
    Yin, Yilong
    IEEE ACCESS, 2018, 6 : 5795 - 5804
  • [40] MOBIL: A Moments based Local Binary Descriptor
    Bellarbi, Abdelkadar
    Otamane, Samir
    Zenati, Nadia
    Benbelkacem, Samir
    2014 IEEE INTERNATIONAL SYMPOSIUM ON MIXED AND AUGMENTED REALITY (ISMAR) - SCIENCE AND TECHNOLOGY, 2014, : 251 - +