| Author: | Liu, Pu |
| Title: | Deep learning for low-resolution human face image recognition |
| Degree: | M.Sc. |
| Year: | 2019 |
| Subject: | Hong Kong Polytechnic University -- Dissertations Human face recognition (Computer science) Image analysis -- Data processing Image processing -- Digital techniques |
| Department: | Faculty of Engineering |
| Pages: | 1, 1, 1, 47 pages : color illustrations |
| Language: | English |
| Abstract: | For the reason that human face images captured by remote monitoring devices are usually of low quality and low resolutions, they are difficult to be accurately recognized in practice. Therefore, this dissertation proposes three different recognition methods for low-resolution human face images. First, we up-sample low-resolution face images and use a conventional eigenface algorithm to do recognition. Then, inspired by the cross-space idea, we adopt the coupled mapping method to project face images of different dimensions into a unified potential subspace, where the recognition can be realized. Finally, we propose a method of deep learning. We use the generated high-resolution images which come from the original low-resolution ones and their corresponding high-resolution images to train our convolutional neural network, and then extract features from the trained model, and the identification is further achieved by the nearest neighborhood method. We evaluate and compare the three proposed methods by using two widely acknowledged datasets. It is proved that the deep learning method has the best performance in low-resolution human face recognition under various conditions. The coupled mapping method outperforms the conventional method. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 991022270856803411.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 1.28 MB | Adobe PDF | View/Open |
Copyright Undertaking
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:
https://theses.lib.polyu.edu.hk/handle/200/10096