DMCA
Face recognition with learning-based Descriptor (2010)
| Venue: | In Proc. IEEE CVPR |
| Citations: | 103 - 13 self |
BibTeX
@INPROCEEDINGS{Cao10facerecognition,
author = {Zhimin Cao and Qi Yin and Jian Sun},
title = {Face recognition with learning-based Descriptor},
booktitle = {In Proc. IEEE CVPR},
year = {2010}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We present a novel approach to address the representa-tion issue and the matching issue in face recognition (verifi-cation). Firstly, our approach encodes the micro-structures of the face by a new learning-based encoding method. Un-like many previous manually designed encoding methods (e.g., LBP or SIFT), we use unsupervised learning tech-niques to learn an encoder from the training examples, which can automatically achieve very good tradeoff be-tween discriminative power and invariance. Then we ap-ply PCA to get a compact face descriptor. We find that a simple normalization mechanism after PCA can further im-prove the discriminative ability of the descriptor. The re-sulting face representation, learning-based (LE) descriptor, is compact, highly discriminative, and easy-to-extract. To handle the large pose variation in real-life scenar-ios, we propose a pose-adaptive matching method that uses pose-specific classifiers to deal with different pose combi-nations (e.g., frontal v.s. frontal, frontal v.s. left) of the matching face pair. Our approach is comparable with the state-of-the-art methods on the Labeled Face in Wild (LFW) benchmark (we achieved 84.45 % recognition rate), while maintaining excellent compactness, simplicity, and gener-alization ability across different datasets.
Keyphrases
face recognition learning-based descriptor pose-specific classifier real-life scenar-ios good tradeoff be-tween discriminative power matching issue gener-alization ability re-sulting face representation novel approach face pair excellent compactness recognition rate labeled face discriminative ability different pose combi-nations simple normalization mechanism large pose variation state-of-the-art method different datasets training example new learning-based encoding method pose-adaptive matching method representa-tion issue compact face descriptor
