A novel feature representation: Aggregating convolution kernels for image retrieval.

Journal: Neural networks : the official journal of the International Neural Network Society
Published Date:

Abstract

Activated hidden units in convolutional neural networks (CNNs), known as feature maps, dominate image representation, which is compact and discriminative. For ultra-large datasets, high dimensional feature maps in float format not only result in high computational complexity, but also occupy massive memory space. To this end, a new image representation by aggregating convolution kernels (ACK) is proposed, where some convolution kernels capturing certain patterns are activated. The top-n index numbers of the convolution kernels are extracted directly as image representation in discrete integer values, which rebuild relationship between convolution kernels and image. Furthermore, a distance measurement is defined from the perspective of ordered sets to calculate position-sensitive similarities between image representations. Extensive experiments conducted on Oxford Buildings, Paris, and Holidays, etc., manifest that the proposed ACK achieves competitive performance on image retrieval with much lower computational cost, outperforming the ones using feature maps for image representation.

Authors

  • Qi Wang
    Biotherapeutics Discovery Research Center, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, 201203, China.
  • Jinxing Lai
    Shaanxi Provincial Major Laboratory for Highway Bridge & Tunnel, Chang'an University, Xi'an 710064, China; School of Highway, Chang'an University, Xi'an 710064, China.
  • Luc Claesen
    Hasselt University, Martelarenlaan 42, Hasselt 3500, Belgium. Electronic address: luc.claesen@uhasselt.be.
  • Zhenguo Yang
    Department of Computer Science, Guangdong University of Technology, Guangzhou, China; Department of Computer Science, City University of Hong Kong, Hong Kong, China. Electronic address: zhengyang5-c@my.cityu.edu.hk.
  • Liang Lei
    Guangdong University of Technology, Guangzhou 510006, China. Electronic address: leiliang@gdut.edu.com.
  • Wenyin Liu
    School of Computer Science and Technology, Guangdong University of Technology, Guangzhou, China. liuwy@gdut.edu.cn.