Deep asymmetric metric learning via rich relationship mining

Xu, Xinyi1; Yang, Yanhua1; Deng, Cheng1; Zheng, Feng2

10.1109/CVPR.2019.00420

2019

1063-6919

978-1-7281-3294-5

32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2019

2019-June

4071-4080

15-20 June 2019

Long Beach, CA, United states

10662 LOS VAQUEROS CIRCLE, PO BOX 3014, LOS ALAMITOS, CA 90720-1264 USA

IEEE Computer Society

Learning effective distance metric between data has gained increasing popularity, for its promising performance on various tasks, such as face verification, zero-shot learning, and image retrieval. A major line of researches employs hard data mining, which makes efforts on searching a subset of significant data. However, hard data mining based approaches only rely on a small percentage of data, which is apt to overfitting. This motivates us to propose a novel framework, named deep asymmetric metric learning via rich relationship mining (DAMLRRM), to mine rich relationship under satisfying sampling size. DAMLRRM constructs two asymmetric data streams that are differently structured and of unequal length. The asymmetric structure enables the two data streams to interlace each other, which allows for the informative comparison between new data pairs over iterations. To improve the generalization ability, we further relax the constraint on the intra-class relationship. Rather than greedily connecting all possible positive pairs, DAMLRRM builds a minimum-cost spanning tree within each category to ensure the formation of a connected region. As such there exists at least one direct or indirect path between arbitrary positive pairs to bridge intra-class relevance. Extensive experimental results on three benchmark datasets including CUB-200-2011, Cars196, and Stanford Online Products show that DAMLRRM effectively boosts the performance of existing deep metric learning approaches.
© 2019 IEEE.

Representation Learning Big Data Large Scale Methods

其他

英语

EI ; CPCI-S

[2017YFE0104100] ; [2018ZDXM-GY-176] ; National Natural Science Foundation of China[61572388]

Computer Science

Computer Science, Artificial Intelligence ; Computer Science, Theory & Methods

WOS:000529484004026

20200508114010

Benchmarking ; Deep learning ; Data mining ; Zero-shot learning

Ergonomics and Human Factors Engineering:461.4 ; Data Processing and Image Processing:723.2

EV Compendex

1.School of Electronic Engineering, Xidian University, Xian; 710071, China
2.Department of Computer Science and Engineering, Southern University of Science and Technology, China

Xu, Xinyi,Yang, Yanhua,Deng, Cheng,et al. Deep asymmetric metric learning via rich relationship mining[C]. 10662 LOS VAQUEROS CIRCLE, PO BOX 3014, LOS ALAMITOS, CA 90720-1264 USA:IEEE Computer Society,2019:4071-4080.