Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Deep Scalable Supervised Quantization by Self-Organizing Map

Deep Scalable Supervised Quantization by Self-Organizing Map Approximate Nearest Neighbor (ANN) search is an important research topic in multimedia and computer vision fields. In this article, we propose a new deep supervised quantization method by Self-Organizing Map to address this problem. Our method integrates the Convolutional Neural Networks and Self-Organizing Map into a unified deep architecture. The overall training objective optimizes supervised quantization loss as well as classification loss. With the supervised quantization objective, we minimize the differences on the maps between similar image pairs and maximize the differences on the maps between dissimilar image pairs. By optimization, the deep architecture can simultaneously extract deep features and quantize the features into suitable nodes in self-organizing map. To make the proposed deep supervised quantization method scalable for large datasets, instead of constructing a larger self-organizing map, we propose to divide the input space into several subspaces and construct self-organizing map in each subspace. The self-organizing maps in all the subspaces implicitly construct a large self-organizing map, which costs less memory and training time than directly constructing a self-organizing map with equal size. The experiments on several public standard datasets prove the superiority of our approaches over the existing ANN search methods. Besides, as a by-product, our deep architecture can be directly applied to visualization with little modification, and promising performance is demonstrated in the experiments. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) Association for Computing Machinery

Deep Scalable Supervised Quantization by Self-Organizing Map

Download PDF
18 pages

Loading next page...
 
/lp/association-for-computing-machinery/deep-scalable-supervised-quantization-by-self-organizing-map-QsvL4Sbz0I
Publisher
Association for Computing Machinery
Copyright
Copyright © 2019 ACM
ISSN
1551-6857
eISSN
1551-6865
DOI
10.1145/3328995
Publisher site
See Article on Publisher Site

Abstract

Approximate Nearest Neighbor (ANN) search is an important research topic in multimedia and computer vision fields. In this article, we propose a new deep supervised quantization method by Self-Organizing Map to address this problem. Our method integrates the Convolutional Neural Networks and Self-Organizing Map into a unified deep architecture. The overall training objective optimizes supervised quantization loss as well as classification loss. With the supervised quantization objective, we minimize the differences on the maps between similar image pairs and maximize the differences on the maps between dissimilar image pairs. By optimization, the deep architecture can simultaneously extract deep features and quantize the features into suitable nodes in self-organizing map. To make the proposed deep supervised quantization method scalable for large datasets, instead of constructing a larger self-organizing map, we propose to divide the input space into several subspaces and construct self-organizing map in each subspace. The self-organizing maps in all the subspaces implicitly construct a large self-organizing map, which costs less memory and training time than directly constructing a self-organizing map with equal size. The experiments on several public standard datasets prove the superiority of our approaches over the existing ANN search methods. Besides, as a by-product, our deep architecture can be directly applied to visualization with little modification, and promising performance is demonstrated in the experiments.

Journal

ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) – Association for Computing Machinery

Published: Aug 20, 2019

Keywords: Approximate nearest neighbor search

References

  • HashGAN: Deep learning to hash with pair conditional wasserstein GAN
    Cao, Yue; Liu, Bin; Long, Mingsheng; Wang, Jianmin; Kliss, MOE
  • Deep cauchy hashing for hamming space retrieval
    Cao, Yue; Long, Mingsheng; Liu, Bin; Wang, Jianmin; Kliss, MOE
  • Deep quantization network for efficient image retrieval
    Cao, Yue; Long, Mingsheng; Wang, Jianmin; Zhu, Han; Wen, Qingfu
  • Return of the devil in the details: Delving deep into convolutional nets
    Chatfield, Ken; Simonyan, Karen; Vedaldi, Andrea; Zisserman, Andrew
  • NUS-WIDE: A real-world web image database from National University of Singapore
    Chua, Tat-Seng; Tang, Jinhui; Hong, Richang; Li, Haojie; Luo, Zhiping; Zheng, Yantao
  • Simultaneous feature aggregating and hashing for large-scale image search
    Do, Thanh-Toan; Tan, Dang-Khoa Le; Pham, Trung T.; Cheung, Ngai-Man
  • Learning deep binary descriptor with multi-quantization
    Duan, Yueqi; Lu, Jiwen; Wang, Ziwei; Feng, Jianjiang; Zhou, Jie
  • Unsupervised deep generative adversarial hashing network
    Dizaji, Kamran Ghasedi; Zheng, Feng; Sadoughi, Najmeh; Yang, Yanhua; Deng, Cheng; Huang, Heng
  • Iterative quantization: A procrustean approach to learning binary codes for large-scale image retrieval
    Gong, Yunchao; Lazebnik, Svetlana; Gordo, Albert; Perronnin, Florent
  • Neural network-based clustering using pairwise constraints
    Hsu, Yen-Chang; Kira, Zsolt
  • Deep image category discovery using a transferred similarity function
    Hsu, Yen-Chang; Lv, Zhaoyang; Kira, Zsolt
  • Collective reconstructive embeddings for cross-modal hashing
    Hu, Mengqiu; Yang, Yang; Shen, Fumin; Xie, Ning; Hong, Richang; Shen, Heng Tao
  • Hashing with angular reconstructive embeddings
    Hu, Mengqiu; Yang, Yang; Shen, Fumin; Xie, Ning; Shen, Heng Tao
  • SUBIC: A supervised, structured binary code for image search
    Jain, Himalaya; Zepeda, Joaquin; Pérez, Patrick; Gribonval, Rémi
  • Product quantization for nearest neighbor search
    Jegou, Herve; Douze, Matthijs; Schmid, Cordelia
  • Asymmetric deep supervised hashing
    Jiang, Qing-Yuan; Li, Wu-Jun
  • Query-adaptive image search with hash codes
    Jiang, Yu-Gang; Wang, Jun; Xue, Xiangyang; Chang, Shih-Fu
  • Self-organized formation of topologically correct feature maps
    Kohonen, Teuvo
  • Kohonen network
    Kohonen, Teuvo; Honkela, Timo
  • Learning multiple layers of features from tiny images
    Krizhevsky, Alex; Hinton, Geoffrey
  • Imagenet classification with deep convolutional neural networks
    Krizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E.
  • Simultaneous feature learning and hash coding with deep neural networks
    Lai, Hanjiang; Pan, Yan; Liu, Ye; Yan, Shuicheng
  • Gradient-based learning applied to document recognition
    LeCun, Yann; Bottou, Léon; Bengio, Yoshua; Haffner, Patrick
  • Deep supervised discrete hashing
    Li, Qi; Sun, Zhenan; He, Ran; Tan, Tieniu
  • Feature learning based deep supervised hashing with pairwise labels
    Li, Wu-Jun; Wang, Sheng; Kang, Wang-Cheng
  • Learning deep parsimonious representations
    Liao, Renjie; Schwing, Alex; Zemel, Richard; Urtasun, Raquel
  • Deep supervised hashing for fast image retrieval
    Liu, Haomiao; Wang, Ruiping; Shan, Shiguang; Chen, Xilin
  • Contextual hashing for large-scale image search
    Liu, Zhen; Li, Houqiang; Zhou, Wengang; Zhao, Ruizhen; Tian, Qi
  • Deep hashing for scalable image search
    Lu, Jiwen; Liong, Venice Erin; Zhou, Jie
  • Robust discrete code modeling for supervised hashing
    Luo, Yadan; Yang, Yang; Shen, Fumin; Huang, Zi; Zhou, Pan; Shen, Heng Tao
  • Cartesian k-means
    Norouzi, Mohammad; Fleet, David J.
  • Overfeat: Integrated recognition, localization and detection using convolutional networks
    Sermanet, Pierre; Eigen, David; Zhang, Xiang; Mathieu, Michaël; Fergus, Rob; LeCun, Yann
  • Deep asymmetric pairwise hashing
    Shen, Fumin; Gao, Xin; Liu, Li; Yang, Yang; Shen, Heng Tao
  • Unsupervised deep hashing with similarity-adaptive and discrete optimization
    Shen, Fumin; Xu, Yan; Liu, Li; Yang, Yang; Huang, Zi; Shen, Heng Tao
  • Matconvnet: Convolutional neural networks for matlab
    Vedaldi, Andrea; Lenc, Karel
  • Learning compact hash codes for multimodal representations using orthogonal deep structure
    Wang, Daixin; Cui, Peng; Ou, Mingdong; Zhu, Wenwu
  • Order preserving hashing for approximate nearest neighbor search
    Wang, Jianfeng; Wang, Jingdong; Yu, Nenghai; Li, Shipeng
  • A general framework for linear distance preserving hashing
    Wang, Min; Zhou, Wengang; Tian, Qi; Li, Houqiang
  • Deep supervised quantization by self-organizing map
    Wang, Min; Zhou, Wengang; Tian, Qi; Pu, Junfu; Li, Houqiang
  • Linear distance preserving pseudo-supervised and unsupervised hashing
    Wang, Min; Zhou, Wengang; Tian, Qi; Zha, Zhengjun; Li, Houqiang
  • Supervised quantization for similarity search
    Wang, Xiaojuan; Zhang, Ting; Qi, Guo-Jun; Tang, Jinhui; Wang, Jingdong
  • Spectral hashing
    Weiss, Yair; Torralba, Antonio; Fergus, Rob
  • Supervised hashing for image retrieval via image representation learning
    Xia, Rongkai; Pan, Yan; Lai, Hanjiang; Liu, Cong; Yan, Shuicheng
  • Generative adversarial product quantisation
    Yu, Litao; Gao, Yongsheng; Zhou, Jun
  • Product quantization network for fast image retrieval
    Yu, Tan; Yuan, Junsong; Fang, Chen; Jin, Hailin
  • Unsupervised deep hashing with pseudo labels for scalable image retrieval
    Zhang, Haofeng; Liu, Li; Long, Yang; Shao, Ling
  • Topology preserving hashing for similarity search
    Zhang, Lei; Zhang, Yongdong; Tang, Jinhui; Gu, Xiaoguang; Li, Jintao; Tian, Qi
  • Bit-scalable deep hashing with regularized similarity learning for image retrieval and person re-identification
    Zhang, Ruimao; Lin, Liang; Zhang, Rui; Zuo, Wangmeng; Zhang, Lei
  • Composite quantization for approximate nearest neighbor search
    Zhang, Ting; Du, Chao; Wang, Jingdong
  • Deep semantic ranking based hashing for multi-label image retrieval
    Zhao, Fang; Huang, Yongzhen; Wang, Liang; Tan, Tieniu
  • BSIFT: Toward data-independent codebook for large scale image search
    Zhou, Wengang; Li, Houqiang; Hong, Richang; Lu, Yijuan; Tian, Qi
  • Scalar quantization for large scale image search
    Zhou, Wengang; Lu, Yijuan; Li, Houqiang; Tian, Qi
  • Towards codebook-free: Scalable cascaded hashing for mobile image search
    Zhou, Wengang; Yang, Ming; Li, Houqiang; Wang, Xiaoyu; Lin, Yuanqing; Tian, Qi
  • Scalable feature matching by dual cascaded scalar quantization for image retrieval
    Zhou, Wengang; Yang, Ming; Wang, Xiaoyu; Li, Houqiang; Lin, Yuanqing; Tian, Qi
  • Locality-constrained deep supervised hashing for image retrieval
    Zhu, Hao; Gao, Shenghua
  • Hash-based block matching for screen content coding
    Zhu, Weijia; Ding, Wenpeng; Xu, Jizheng; Shi, Yunhui; Yin, Baocai