Skip to main navigation Skip to search Skip to main content

Saliency meets spatial quantization: A practical framework for large scale product search

  • Harbin Institute of Technology Shenzhen
  • National University of Singapore
  • CAS - Institute of Computing Technology
  • PKU-HKUST Shenzhen-Hong Kong Institution
  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Product image search aims to retrieve similar product images based on a query image. While deep learning based features work well in retrieving images of the same category (e.g. 'searching for T-shirts from all the clothing images'), they perform poorly when retrieving variants of images within the same category (e.g. 'searching for uniform of Chelsea football club from all T-shirts image'), since it requires fine grained matching on image details. In this paper, we present a spatial quantization approach that utilizes spatial pyramid pooling (SPP) and vector of locally aggregated descriptors (VLAD) to extract more discriminative features for style-aware product search. By using the proposed spatial quantization, spatial information is encoded into the image feature to improve the fine grained product image search. Finally, the experiments on a large scale real world dataset provided by Alibaba large-scale image search challenge (ALISC) demonstrate the effectiveness of our method.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509015528
DOIs
StatePublished - 22 Sep 2016
Externally publishedYes
Event2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016 - Seattle, United States
Duration: 11 Jul 201615 Jul 2016

Publication series

Name2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016

Conference

Conference2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016
Country/TerritoryUnited States
CitySeattle
Period11/07/1615/07/16

Keywords

  • Image retrieval
  • Salient region detection
  • Vector quantization

Fingerprint

Dive into the research topics of 'Saliency meets spatial quantization: A practical framework for large scale product search'. Together they form a unique fingerprint.

Cite this