Submodular Object Recognition

Zhu, Fan, Jiang, Zhuolin and Shao, Ling (2014) Submodular Object Recognition. In: CVPR 2014 - IEEE Conference on Computer Vision and Pattern Recognition, 23rd - 28th June 2014, Columbus, Ohio.

Full text not available from this repository. (Request a copy)

Abstract

We present a novel object recognition framework based on multiple figure-ground hypotheses with a large object spatial support, generated by bottom-up processes and mid-level cues in an unsupervised manner. We exploit the benefit of regression for discriminating segments' categories and qualities, where a regressor is trained to each category using the overlapping observations between each figure-ground segment hypothesis and the ground-truth of the target category in an image. Object recognition is achieved by maximizing a submodular objective function, which maximizes the similarities between the selected segments (i.e., facility locations) and their group elements (i.e., clients), penalizes the number of selected segments, and more importantly, encourages the consistency of object categories corresponding to maximum regression values from different category-specific regressors for the selected segments. The proposed framework achieves impressive recognition results on three benchmark datasets, including PASCAL VOC 2007, Caltech-101 and ETHZ-shape.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	G400 Computer Science
Department:	Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User:	Paul Burns
Date Deposited:	16 Jun 2015 09:27
Last Modified:	13 Oct 2019 00:37
URI:	http://nrl.northumbria.ac.uk/id/eprint/22927

Downloads