CASIA OpenIR  > 智能感知与计算研究中心
X-GACMN: An X-Shaped Generative Adversarial Cross-Modal Network with Hypersphere Embedding
Weikuo Guo1; Jian Liang2,3; Xiangwei Kong1; Lingxiao Song2; Ran He2,3
2018
Conference NameAsian Conference on Computer Vision (ACCV 2018)
Conference Date2018
Conference PlaceAustralia
Abstract

How to bridge heterogeneous gap between different modalities is one of the main challenges in cross-modal retrieval task. Most existing methods try to tackle this problem by projecting data from different modalities into a common space. In this paper, we introduce a novel X-Shaped Generative Adversarial Cross-Modal Network (X-GACMN) to learn a better common space between different modalities. Specifically, the proposed architecture combines the process of synthetic data generation and distribution adapting into a unified framework to make sure the heterogeneous modality distributions similar to each other in the learned common subspace. To promote the discriminative ability, a new loss function that combines intra-modality angular softmax loss and cross-modality pair-wise consistent loss is further imposed on the common space, hence the learned features can well preserve both intermodality structure and intra-modality structure on a hypersphere manifold. Extensive experiments on three benchmark datasets show the effectiveness of the proposed approach.

Indexed ByEI
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/23807
Collection智能感知与计算研究中心
Corresponding AuthorXiangwei Kong
Affiliation1.Dalian University of Technology
2.University of Chinese Academy of Science(UCAS)
3.CRIPAC and NLPR, CASIA
Recommended Citation
GB/T 7714
Weikuo Guo,Jian Liang,Xiangwei Kong,et al. X-GACMN: An X-Shaped Generative Adversarial Cross-Modal Network with Hypersphere Embedding[C],2018.
Files in This Item: Download All
File Name/Size DocType Version Access License
accv18.pdf(1088KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Weikuo Guo]'s Articles
[Jian Liang]'s Articles
[Xiangwei Kong]'s Articles
Baidu academic
Similar articles in Baidu academic
[Weikuo Guo]'s Articles
[Jian Liang]'s Articles
[Xiangwei Kong]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Weikuo Guo]'s Articles
[Jian Liang]'s Articles
[Xiangwei Kong]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: accv18.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.