Please use this identifier to cite or link to this item: https://doi.org/10.1109/CVPR.2012.6248094
Title: Omni-range spatial contexts for visual classification
Authors: Ni, B.
Xu, M.
Tang, J.
Yan, S. 
Moulin, P.
Issue Date: 2012
Source: Ni, B.,Xu, M.,Tang, J.,Yan, S.,Moulin, P. (2012). Omni-range spatial contexts for visual classification. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition : 3514-3521. ScholarBank@NUS Repository. https://doi.org/10.1109/CVPR.2012.6248094
Abstract: Spatial contexts encode rich discriminative information for visual classification. However, as object shapes and scales vary significantly among images, spatial contexts with manually specified distance ranges are not guaranteed with optimality. In this work, we investigate how to automatically select discriminative and stable distance bin groups for modeling image spatial contexts to improve classification performance. We make two observations. First, the number of distance bins for context modeling can be arbitrarily large, and discriminative contexts are only from a small subset of distance bins. Second, adjacent distance bins for contexts modeling often show similar characteristics, thus encouraging grouping them together can result in more stable representation. Utilizing these two observations, we propose an omni-range spatial context mining framework for image classification. A sparse selection and grouping regularizer is employed along with an empirical risk, to discover discriminative and stable distance bin groups for context modeling. To facilitate efficient optimization, the objective function is approximated by a smooth convex function with theoretically guaranteed error bounds. The selected and grouped image spatial contexts, which are applied in food and national flag recognition, are demonstrated to be discriminative, compact and robust. © 2012 IEEE.
Source Title: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
URI: http://scholarbank.nus.edu.sg/handle/10635/71200
ISBN: 9781467312264
ISSN: 10636919
DOI: 10.1109/CVPR.2012.6248094
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

5
checked on Jan 9, 2018

Page view(s)

27
checked on Jan 19, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.