Illumination ratio image: synthesizing and recognition with varying illuminations

被引：25

作者：

Zhao, JL ^{[1
]}

Su, Y

Wang, DJ

Luo, SW

机构：

[1] Samsung Adv Inst Technol, Beijing Lab Electron Microscopy, Beijing 100081, Peoples R China

[2] CASCO Corp, Beijing 100055, Peoples R China

[3] No Jiaotong Univ, Dept Comp Sci, Beijing 100044, Peoples R China

来源：

PATTERN RECOGNITION LETTERS | 2003年 / 24卷 / 15期

关键词：

face recognition; illumination modeling; face synthesis; image-based rendering;

D O I：

10.1016/S0167-8655(03)00113-2

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Illumination ratio images are presented in the paper to illuminate the faces. Our rendering method can simulate the distribution of the images with varying illuminations and generate new training images for face recognition with single frontal view. We analyzed the reason of the assumption that different persons have the same surface normal but different albedo. The synthesized images were compared with the original images by Gamma correction and error images. Our experiment shows that eyes are not lambertian surfaces. The synthesized images improve the face recognition performance by using the individual eigenface classifier. (C) 2003 Elsevier B.V. All rights reserved.

引用

页码：2703 / 2710

页数：8

共 12 条

[1]

[Anonymous], P IEEE C CVPR

[2]

[Anonymous], 3 WORKSH EMP EV METH

[3]

Basri R, 2001, EIGHTH IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, VOL II, PROCEEDINGS, P383, DOI 10.1109/ICCV.2001.937651

[4]

Belhumeur P. N., 1997, P IEEE C COMP VIS PA, P52

[5]

Georghiades A. S., 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580), P277, DOI 10.1109/AFGR.2000.840647

[6] Illumination cones for recognition under variable lighting: Faces [J].

Georghiades, AS ;

Kriegman, DJ ;

Belhumeur, PN .

1998 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, 1998, :52-58

[7]

HALLINAN PW, 1994, 1994 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, P995, DOI 10.1109/CVPR.1994.323941

[8]

MOSES Y, 1994, EUR C COMP VIS, P286

[9]

NAYAR S, 1994, CUCS02194

[10] The quotient image: Class-based re-rendering and recognition with varying illuminations [J].

Shashua, A ;

Riklin-Raviv, T .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2001, 23 (02) :129-139

← 1 2 →