Learning to Match Clothing From Textual Feature-Based Compatible Relationships

Learning to Match Clothing From Textual Feature-Based Compatible Relationships
复制标题

从基于文本特征的兼容关系中学习匹配服装

DOI:
10.1109/tii.2019.2924725
复制
发表时间:
2020-11
影响因子:
12.3
通讯作者:
Tommy W.S. Chow
Tommy W.S. Chow
中科院分区:
计算机科学1区
文献类型:
--
作者:
Haijun Zhang;Wang Huang;Linlin Liu;Tommy W.S. Chow

文献摘要

参考文献

被引文献

相似文献

本文提出了一种考虑物品间兼容性的服装匹配新框架。与使用服装的视觉特征相比,我们只使用它们的文字描述,即标题句来构成基本特征。其中,长短期记忆(LSTM)网络用于标题句的特征嵌入。给定查询和候选项对,将Siamese lstm实现的特征嵌入集成到以兼容矩阵为特征的风格兼容空间中。我们的框架分别在亚马逊、淘宝和Polyvore三个大型服装项目集上进行了检验。实验证实了该方法与几种基线方法的有效性。
This paper presents a new framework for matching clothes by considering item in-between compatibility. In contrast to the use of visual features of clothing items, we only utilized their textual descriptions, i.e., title sentences, to constitute the basic features. Specifically, a longshort-term memory (LSTM) network was used for feature embeddings of title sentences. Given item pairs of queries and candidates, their feature embeddings achieved by Siamese LSTMs were integrated into style-compatible space characterized by a compatibility matrix. Our framework is examined on three large-scaled clothing item sets collected from Amazon, Taobao, and Polyvore, respectively. Experiments confirm the efficacy of our approach compared with several baseline methods.
DOI: 10.1109/tkde.2005.50
发表时间: 2005-03-01
影响因子: 8.9
作者:
Huang, J;Ling, CX
通讯作者: Ling, CX
DOI: 10.1109/tcyb.2016.2593940
发表时间: 2017-11-01
影响因子: 11.8
作者:
Gonzalez, Alejandro;Vazquez, David;Amores, Jaume
通讯作者: Amores, Jaume
DOI: 10.5591/978-1-57735-516-8/ijcai11-377
发表时间: 2011-07
期刊: --
影响因子: --
作者:
Tomoharu Iwata;Shinji Watanabe;H. Sawada
通讯作者: Tomoharu Iwata;Shinji Watanabe;H. Sawada
DOI: 10.7766/orbit.v1.2.44
发表时间: 2001-01
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
D. Blei;Andrew Y. Ng;Michael I. Jordan
通讯作者: D. Blei;Andrew Y. Ng;Michael I. Jordan
DOI: --
发表时间: 2017-08
期刊: ArXiv
影响因子: --
作者:
Kui Zhao;Xia Hu;Jiajun Bu;Can Wang
通讯作者: Kui Zhao;Xia Hu;Jiajun Bu;Can Wang