A review of image set classification

In computer vision, we generally solve a classification problem by a single image. With the video cameras being widely used in our real life, it is a nature choice to solve a classification problem by image sets. Compared with the single image based methods, the image set classification deals with s...

Full description

Saved in:
Bibliographic Details
Published in:Neurocomputing (Amsterdam) 2019-03, Vol.335, p.251-260
Main Authors: Zhao, Zhong-Qiu, Xu, Shou-Tao, Liu, Dian, Tian, Wei-Dong, Jiang, Zhi-Da
Format: Article
Language:English
Subjects:
Affine hull
Face recognition
Image set classification
Linear subspace
Nonlinear manifold
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In computer vision, we generally solve a classification problem by a single image. With the video cameras being widely used in our real life, it is a nature choice to solve a classification problem by image sets. Compared with the single image based methods, the image set classification deals with severe changes of appearance and makes decisions by comparing the query set with gallery sets. So the image set classification offers more promises and has therefore attracted significant research attention in recent years. In this paper, we provide a review on image set classification. Our review begins with an overview of the direction of image set classification. Then we detail some classic algorithms. Experimental analyses are provided in corresponding subsection to compare classification performance of various methods and draw some meaningful conclusions. Finally, several promising directions and tasks are provided as guidelines for future work.
ISSN:0925-2312
1872-8286
DOI:10.1016/j.neucom.2018.09.090