Class specific subspace dependent nonlinear correlation filtering for illumination tolerant face recognition

Authors:
Pradipta K. Banerjee;Asit K. Datta
Affiliations:
-;-
Venue:
Pattern Recognition Letters
Year:
2014

Citing 7
Cited 0

The CMU Pose, Illumination, and Expression Database

IEEE Transactions on Pattern Analysis and Machine Intelligence
Acquiring Linear Subspaces for Face Recognition under Variable Lighting

IEEE Transactions on Pattern Analysis and Machine Intelligence
Face recognition subject to variations in facial expression, illumination and pose using correlation filters

Computer Vision and Image Understanding
Eigenfeature Regularization and Extraction in Face Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Illumination compensation and normalization in eigenspace-based face recognition: A comparative study of different pre-processing approaches

Pattern Recognition Letters
Reliable face recognition using adaptive and robust correlation filters

Computer Vision and Image Understanding
The correntropy MACE filter

Pattern Recognition

Quantified Score

Hi-index	0.10

Visualization

Abstract

A frequency domain nonlinear correlation technique for face recognition under varying lighting conditions is proposed. The technique is based on phase correlation between an optimum projecting image correlation filter and an optimum reconstructed image correlation filter during class specific subspace operation. Performance improvement is achieved by exploiting point wise nonlinearities of image pixels. Further optimization is carried out by minimizing the energy at the correlation plane while maximizing the correlation peak. While comparing with other standard unconstrained correlation filters, improved performance of the proposed scheme is established by experimental results on standard face data bases.