Revisiting image ordinal estimation: how to deal with ordinal relationship in deep learning?

Chao Zhang; Ce Zhu; Xun Xu

doi:10.1117/1.JEI.28.1.013025

5 February 2019 Revisiting image ordinal estimation: how to deal with ordinal relationship in deep learning?

Chao Zhang, Ce Zhu, Xun Xu

Author Affiliations +

Journal of Electronic Imaging, Vol. 28, Issue 1, 013025 (February 2019). https://doi.org/10.1117/1.JEI.28.1.013025

Abstract

Image ordinal estimation usually defined as a regression or classification problem is challenging in computer vision. Many recent works simply implement deep convolutional neural networks (CNNs) to get ideal results, while they have not analyzed why it works well as conventional computer vision methods. We revisit the problem of image ordinal estimation and propose two methods on ordinal estimation. First, a double-task convolutional neural network model (DTCNN) is proposed by combining classification and regression objectives. Then we have analyzed why DTCNN works well from three perspectives: (1) the ordinal relationship between two tasks, (2) the coarse/fine category level for two tasks, and (3) neurons’ activation on two tasks. To avoid choosing heuristic combination parameter, we further propose risk-based CNN (R-CNN) model to embed the ordinal constraint into classification task. Based on Bayesian risk rule, the category information and ordinal information are fused by a dynamic weighting loss function. Experimental results on two ordinal datasets show that both DTCNN and R-CNN obtain competitive performance.

Citation Download Citation

Chao Zhang, Ce Zhu, and Xun Xu "Revisiting image ordinal estimation: how to deal with ordinal relationship in deep learning?," Journal of Electronic Imaging 28(1), 013025 (5 February 2019). https://doi.org/10.1117/1.JEI.28.1.013025

Received: 19 October 2018; Accepted: 9 January 2019; Published: 5 February 2019

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