Abstract
This paper presents a computational model to address one prominent psychological behavior of human beings to recognize images. The basic pursuit of our method can be concluded as that differences among multiple images help visual recognition. Generally speaking, we propose a statistical framework to distinguish what kind of image features capture sufficient category information and what kind of image features are common ones shared in multiple classes. Mathematically, the whole formulation is subject to a generative probabilistic model. Meanwhile, a discriminative functionality is incorporated into the model to interpret the differences among all kinds of images. The whole Bayesian formulation is solved in an Expectation-Maximization paradigm. After finding those discriminative patterns among different images, we design an image categorization algorithm to interpret how these differences help visual recognition within the bag-of-feature framework. The proposed method is verified on a variety of image categorization tasks including outdoor scene images, indoor scene images as well as the airborne SAR images from different perspectives.
Highlights
Visual information understanding is a long-standing topic that has been extensively discussed in both the communities of vision research and artificial intelligence
We will report some numerical properties of the statistical model and two experiments on bi-class and multi-class classification will be respectively conducted to verify the effectiveness of the selected discriminative features for image categorization
Data Collection The data are from three datasets, Synthetic Aperture Radar (SAR) dataset, Fifteen Scene dataset and MIT Indoor Scene dataset, among which SAR dataset is established by us
Summary
Visual information understanding is a long-standing topic that has been extensively discussed in both the communities of vision research and artificial intelligence. When distinguishing two images, we pay much attention to their differences rather than the common characters. When categorizing two images from bedroom category and office category, the most representative information that helps recognition are the ‘‘beds’’ and ‘‘computers’’. When finding the features of a computer in an image, it probably describes an office scenario. In both of these two images, other common patterns are collected, e.g. the wall and the ground. These common features help less to distinguish these two images because they appear in both of them
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
