Topic network: topic model with deep learning for image classification

Zhiyong Pan; Yang Liu; Guojun Liu; Maozu Guo; Yang Li

doi:10.1117/1.JEI.27.3.033009

12 May 2018 Topic network: topic model with deep learning for image classification

Zhiyong Pan, Yang Liu, Guojun Liu, Maozu Guo, Yang Li

Author Affiliations +

Journal of Electronic Imaging, Vol. 27, Issue 3, 033009 (May 2018). https://doi.org/10.1117/1.JEI.27.3.033009

Abstract

As a representative deep learning model, convolutional neural networks (CNNs) have accomplished great achievements in image classification and object detection. However, CNNs require the resizing of the input images to a fixed size, which may affect the representations of objects. To overcome this limitation, we replace the last pooling layer with a topic model and call it a topic network. For arbitrary sizes and ratios of input images, the outputs of the topic network are fixed-size features due to the topic model of the topic layer, and they can reflect the global or regional characteristics of images by means of different scales. Two topic models, namely latent Dirichlet allocation (LDA) and Markov topic random fields (MTRF), are applied to the topic layer, and we call them latent Dirichlet allocation topic network and Markov topic random fields topic network, respectively. Both of them perform well in image classification with original size. More importantly, as a framework, any topic model can be easily applied to the topic layer of a topic network, which makes it much more flexible and extensible.

Citation Download Citation

Zhiyong Pan, Yang Liu, Guojun Liu, Maozu Guo, and Yang Li "Topic network: topic model with deep learning for image classification," Journal of Electronic Imaging 27(3), 033009 (12 May 2018). https://doi.org/10.1117/1.JEI.27.3.033009

Received: 12 December 2017; Accepted: 12 April 2018; Published: 12 May 2018

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