Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Multi-disciplinary Studies | en_US |
| dc.contributor | Department of Electronic Engineering | en_US |
| dc.creator | Liu, Shu-tim | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/1561 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | - |
| dc.rights | All rights reserved | en_US |
| dc.title | Surface reconstruction of human face from stereo images | en_US |
| dcterms.abstract | Eigenfaces for recognition has advantages over other machine recognition schemes of face in its speed and simplicity, learning capacity, and relative insensitivity to small or gradual changes in the face image. Moreover, this scheme has practically been tested on a large database. However, the recognition rate decreases as the difference of viewing angle between a face image and its corresponding reference stored in the database increases. This is because conventional algorithms recognize a face by extracting the relevant information in a face image, encoding it with eigenfaces, and comparing its encoded output against a database of identified encoded faces. Thus, we proposed a new scheme to improve this weakness. In stead of performing the face recognition in 2D domain, we compare the depth maps of faces after performing a three-dimensional reconstruction of stereo images. Our approach makes use of a pair of stereo images to recover its disparity map by applying a parallel stereo matching algorithm that produces dense depth maps and preserves image features. The disparity map can pass through a geometrical calculation of stereo cameras to locate a reference 3D coordination system then. Since the obtained depth maps are typically noisy and not fully populated with known data points due to false match points and lack of texture, a mathematical curve fitting is necessary to fill in missing points and smooth the data. An adaptive smoothing algorithm has been investigated and used to refine the depth map and reconstruct 3D surfaces. This surface reconstruction scheme can be applied to a wide variety of problems, including face recognition, criminal identification, security systems and computer vision. | en_US |
| dcterms.extent | vi, 71 leaves : ill. (some col.) ; 30 cm | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 1999 | en_US |
| dcterms.educationalLevel | All Master | en_US |
| dcterms.educationalLevel | M.Sc. | en_US |
| dcterms.LCSH | Human face recognition (Computer science) | en_US |
| dcterms.LCSH | Computer vision | en_US |
| dcterms.LCSH | Image processing | en_US |
| dcterms.LCSH | Image reconstruction | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b14854107.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 4.45 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/200/1561