Clothing-to-words mapping using word separation method

  • Authors:

  • Jinglei Zhou;Mao Ye;Jian Ding;Haiyang Wang;Xue Li

  • Affiliations:

  • School of Economics and Management, Southwest Jiaotong University, Chengdu 610031, PR China;School of Economics and Management, Southwest Jiaotong University, Chengdu 610031, PR China and School of Computer Science and Engineering, University of Electronic Science and Technology of China ...;School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, PR China;School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, PR China;School of Information Technology and Electrical Engineering, The University of Queensland, Brisbane, Queensland 4072, Australia

  • Venue:
  • Computers and Electrical Engineering
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

With the development of E-commerce, clothing search on Internet emerges to be a valuable and challenging problem. Compared with the standard image retrieval approach, there are two main difficulties in clothing search. The first is the numerous clothing variation. Another is that people like to search the clothing, which have the same visual elements under the numerous variation. Motivated by Graph Cut method, an approach called word separation method is proposed to map the clothing visual elements to words, which can simultaneously take into account the image-to-image relationship, the image-to-word relationship and the word-to-word relationship. In our work, the meaningful words from web pages are represented by the graph nodes. The graph edges are weighted by the context of data set, which is from Internet. The experimental results on the clothing data set demonstrate the efficiency, effectiveness and robustness of our method.