City Research Online

Abstract

In this paper, we propose convex cone-based frameworks for image-set classification. Image-set classification aims to classify a set of images, usually obtained from video frames or multi-view cameras, into a target object. To accurately and stably classify a set, it is essential to accurately represent structural information of the set. There are various image features, such as histogram-based features and convolutional neural network features. We should note that most of them have non-negativity and thus can be effectively represented by a convex cone. This leads us to introduce the convex cone representation to image-set classification. To establish a convex cone-based framework, we mathematically define multiple angles between two convex cones, and then use the angles to define the geometric similarity between them. Moreover, to enhance the framework, we introduce two discriminant spaces. We first propose a discriminant space that maximizes gaps between cones and minimizes the within-class variance. We then extend it to a weighted discriminant space by introducing weights on the gaps to deal with complicated data distribution. In addition, to reduce the computational cost of the proposed methods, we develop a novel strategy for fast implementation. The effectiveness of the proposed methods is demonstrated experimentally by using five databases.

Publication Type:	Article
Additional Information:	© 2021. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ . This article has been published in Pattern Recognition, DOI: https://doi.org/10.1016/j.patcog.2021.108190
Publisher Keywords:	Image-set based method, Convex cone representation, Multiple angles
Subjects:	Q Science > QA Mathematics T Technology > T Technology (General)
Departments:	Bayes Business School > Actuarial Science & Insurance
SWORD Depositor:	Symplectic Administrator

Export

Downloads