|Title:||Low-resolution face recognition|
|Subject:||Hong Kong Polytechnic University -- Dissertations|
Human face recognition (Computer science)
|Department:||Department of Electronic and Information Engineering|
|Pages:||31, iv pages : color illustrations|
|Abstract:||Face recognition has got rapid development during this decade. However, low-resolution face recognition usually cannot achieve satisfactory performance, whether by using traditional methods or deep learning methods. In general, the amount of information and the usefulness of features will determine the face recognition rate. In this dissertation, after reviewing some of the previous works, we propose a practical multi-resolution method namely "MB-LBPD-MDS". Firstly, MTCNN is employed for the face detection stage. Secondly, for feature extraction, we have implemented "MB-LBPD" to make the Local Binary Pattern features become numerical so that it can be used to compute the covariance matrix for principal component analysis. Then, we use the MDS (multidimensional scaling) method to simultaneously project high-resolution (HR) images and low-resolution (LR) images into a unified latent subspace, where the distances of different classes approximate their distances in the HR space. The proposed approach is able to separate interclass faces accurately, because, the optimization iteration algorithm used can not only ensure the consistency for each LR face image and corresponding HR one, but also take both intraclass distances and interclass distances into consideration. Experiments on public face databases, like the ORL database and FERET database, show satisfactory results, and a large number of experiments have also been conducted in this project.|
Files in This Item:
|991022270855803411.pdf||For All Users (off-campus access for PolyU Staff & Students only)||675.77 kB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: