A similarity learning for fine-grained images based on the Mahalanobis metric and the kernel method

Zisheng Fu; Ninghua Wang; Zhimin Feng; Ting Dong

doi:10.1117/12.2501757

26 July 2018 A similarity learning for fine-grained images based on the Mahalanobis metric and the kernel method

Zisheng Fu, Ninghua Wang, Zhimin Feng, Ting Dong

Proceedings Volume 10828, Third International Workshop on Pattern Recognition; 1082815 (2018) https://doi.org/10.1117/12.2501757
Event: Third International Workshop on Pattern Recognition, 2018, Jinan, China

Abstract

Since most prior studies on similar image retrieval focused on the category level, image similarity learning at the finegrained level remains challenge, which often leads to a semantic gap between the low-level visual features and highlevel human perception. To solve the problem, we proposed a Mahalanobis and kernel-based similarity (Mah-Ker) method combined with features developed by the Convolutional Neural Network (CNN). Firstly, triplet constraints are introduced to characterize the fine-grained image similarity relationship which the Mahalanobis metric is trained upon. Then a kernel-based metric is proposed in the last layer of model to devise nonlinear extensions of Mahalanobis metric and further enhance the performance. Experiments based on the real VIP.com dress dataset showed that our proposed method achieved a promising higher retrieval performance than both the state-of-art fine-grained similarity model and the hand-crafted visual feature based approaches.

Citation Download Citation

Zisheng Fu, Ninghua Wang, Zhimin Feng, and Ting Dong "A similarity learning for fine-grained images based on the Mahalanobis metric and the kernel method", Proc. SPIE 10828, Third International Workshop on Pattern Recognition, 1082815 (26 July 2018); https://doi.org/10.1117/12.2501757

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available