An Adversarial Learning and Canonical Correlation Analysis Based Cross-Modal Retrieval Model

被引：0

作者：

Thi-Hong Vuong ^{[1
]}

Thanh-Huyen Pham ^{[1
,2
]}

Tri-Thanh Nguyen ^{[1
]}

Quang-Thuy Ha ^{[1
]}

机构：

[1] UET, VNU, Hanoi VNU, 144 Xuan Thuy, Hanoi, Vietnam

[2] Ha Long Univ, Quang Ninh, Vietnam

来源：

INTELLIGENT INFORMATION AND DATABASE SYSTEMS, ACIIDS 2019, PT I | 2019年 / 11431卷

关键词：

Cross-modal retrieval; Adversarial learning; Canonical correlation analysis;

D O I：

10.1007/978-3-030-14799-0_13

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The key of cross-modal retrieval approaches is to find a maximally correlated subspace among multiple datasets. This paper introduces a novel Adversarial Learning and Canonical Correlation Analysis based Cross-Modal Retrieval (ALCCA-CMR) model. For each modality, the ALCCA phase finds an effective common subspace and calculates the similarity by canonical correlation analysis embedding for cross-modal retrieval. We demonstrate an application of ALCCA-CMR model implemented for the dataset of two modalities. Experimental results on real music data show the efficacy of the proposed method in comparison with other existing ones.

引用

页码：153 / 164

页数：12

共 22 条

[1] Andrienko G., 2013, Introduction, P1
[2] [Anonymous], ARXIV171108976
[3] [Anonymous], 2014, AAAI
[4] Photo classification by integrating image content and camera metadata
Boutell, M
Luo, JB
[J]. PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 4, 2004, : 901 - 904
[5] Chaudhuri Kamalika, 2009, P 26 ANN INT C MACH, P129
[6] De Bie T., 2003, P INT S IND COMP AN, P785
[7] Deep correspondence restricted Boltzmann machine for cross-modal retrieval
Feng, Fangxiang
Li, Ruifan
Wang, Xiaojie
[J]. NEUROCOMPUTING, 2015, 154 : 50 - 60
[8] Goodfellow IJ, 2014, ADV NEUR IN, V27, P2672
[9] Hu X, 2009, 10 INT SOC MUS INF R, P411
[10] Le Q., 2014, ICML, P1188

← 1 2 3 →