Face identification across different poses and illuminations with a 3D morphable model

被引：126

作者：

Blanz, V ^{[1
]}

Romdhani, S ^{[1
]}

Vetter, T ^{[1
]}

机构：

[1] Univ Freiburg, D-79110 Freiburg, Germany

来源：

FIFTH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE AND GESTURE RECOGNITION, PROCEEDINGS | 2002年

关键词：

D O I：

10.1109/AFGR.2002.1004155

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 [模式识别与智能系统]; 0812 [计算机科学与技术]; 0835 [软件工程]; 1405 [智能科学与技术];

摘要：

We present a novel approach for recognizing faces in images taken from different directions and under different illumination. The method is based on a 3D morphable face model that encodes shape and texture in terms of model parameters, and an algorithm that recovers these parameters from a single image of a face. For face identification, we use the shape and texture parameters of the model that are separated from imaging parameters, such as pose and illumination. In addition to the identity, the system provides a measure of confidence. We report experimental results for more than 4000 images from the publicly available CMU-PIE database.

引用

页码：202 / 207

页数：6

共 15 条

[1]

[Anonymous], 2001, NEURAL NETWORKS COMP

[2]

BERGEN JR, 1990, HIERARCHICAL MOTION

[3]

A morphable model for the synthesis of 3D faces [J].

Blanz, V ;

Vetter, T .

SIGGRAPH 99 CONFERENCE PROCEEDINGS, 1999, :187-194

[4]

Cootes T. F., 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580), P227, DOI 10.1109/AFGR.2000.840639

[5]

Foley JD, 1984, FUNDAMENTALS INTERAC

[6]

From few to many: Illumination cone models for face recognition under variable lighting and pose [J].