Principal component analysis of image gradient orientations for face recognition

We introduce the notion of Principal Component Analysis (PCA) of image gradient orientations. As image data is typically noisy, but noise is substantially different from Gaussian, traditional PCA of pixel intensities very often fails to estimate reliably the low-dimensional subspace of a given data...

Full description

Bibliographic Details
Main Authors: Tzimiropoulos, Georgios, Zafeiriou, Stefanos, Pantic, Maja
Format: Conference or Workshop Item
Published: 2011
Subjects:
Online Access:https://eprints.nottingham.ac.uk/31408/
_version_ 1848794195080249344
author Tzimiropoulos, Georgios
Zafeiriou, Stefanos
Pantic, Maja
author_facet Tzimiropoulos, Georgios
Zafeiriou, Stefanos
Pantic, Maja
author_sort Tzimiropoulos, Georgios
building Nottingham Research Data Repository
collection Online Access
description We introduce the notion of Principal Component Analysis (PCA) of image gradient orientations. As image data is typically noisy, but noise is substantially different from Gaussian, traditional PCA of pixel intensities very often fails to estimate reliably the low-dimensional subspace of a given data population. We show that replacing intensities with gradient orientations and the ℓ2 norm with a cosine-based distance measure offers, to some extend, a remedy to this problem. Our scheme requires the eigen-decomposition of a covariance matrix and is as computationally efficient as standard ℓ2 intensity-based PCA. We demonstrate some of its favorable properties for the application of face recognition.
first_indexed 2025-11-14T19:12:20Z
format Conference or Workshop Item
id nottingham-31408
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:12:20Z
publishDate 2011
recordtype eprints
repository_type Digital Repository
spelling nottingham-314082020-05-04T20:23:29Z https://eprints.nottingham.ac.uk/31408/ Principal component analysis of image gradient orientations for face recognition Tzimiropoulos, Georgios Zafeiriou, Stefanos Pantic, Maja We introduce the notion of Principal Component Analysis (PCA) of image gradient orientations. As image data is typically noisy, but noise is substantially different from Gaussian, traditional PCA of pixel intensities very often fails to estimate reliably the low-dimensional subspace of a given data population. We show that replacing intensities with gradient orientations and the ℓ2 norm with a cosine-based distance measure offers, to some extend, a remedy to this problem. Our scheme requires the eigen-decomposition of a covariance matrix and is as computationally efficient as standard ℓ2 intensity-based PCA. We demonstrate some of its favorable properties for the application of face recognition. 2011-03 Conference or Workshop Item PeerReviewed Tzimiropoulos, Georgios, Zafeiriou, Stefanos and Pantic, Maja (2011) Principal component analysis of image gradient orientations for face recognition. In: 2011 IEEE International Conference on Automatic Face & Gesture Recognition (FG 2011), 21-25 March 2011, Santa Barbara, California, USA. Face recognition Gradient methods Principal component analysis http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5771457&punumber%3D5765597%26filter%3DAND%28p_IS_Number%3A5771322%29%26pageNumber%3D4
spellingShingle Face recognition
Gradient methods
Principal component analysis
Tzimiropoulos, Georgios
Zafeiriou, Stefanos
Pantic, Maja
Principal component analysis of image gradient orientations for face recognition
title Principal component analysis of image gradient orientations for face recognition
title_full Principal component analysis of image gradient orientations for face recognition
title_fullStr Principal component analysis of image gradient orientations for face recognition
title_full_unstemmed Principal component analysis of image gradient orientations for face recognition
title_short Principal component analysis of image gradient orientations for face recognition
title_sort principal component analysis of image gradient orientations for face recognition
topic Face recognition
Gradient methods
Principal component analysis
url https://eprints.nottingham.ac.uk/31408/
https://eprints.nottingham.ac.uk/31408/