Clothing Retrieval Based on Local Similarity with Multiple Images

被引：12

作者：

Mizuochi, Masaru ^{[1
]}

Kanezaki, Asako ^{[1
]}

Harada, Tatsuya ^{[1
]}

机构：

[1] Univ Tokyo, Tokyo, Japan

来源：

PROCEEDINGS OF THE 2014 ACM CONFERENCE ON MULTIMEDIA (MM'14) | 2014年

关键词：

garment image search; feature vector coding; local descriptors; user interaction; system design;

D O I：

10.1145/2647868.2655021

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

Recently, the online shopping market has been expanded, which has advanced studies of clothing retrieval via image search. For this study, we develop a novel clothing retrieval system considering local similarity, where users can retrieve their desired clothes which are globally similar to an image and partially similar to another image. We propose a method of coding global features by merging local descriptors extracted from multiple images. Furthermore, we design a system that re-evaluates output of similar image search by the similarity of local regions. We demonstrated that our method increased the probability of users finding their desired clothes from 39.7%-55.1%, compared to a standard similar image search system with global features of a single image. Statistical significance is proven using t-tests.

引用

页码：1165 / 1168

页数：4