Toggle Main Menu Toggle Search

Open Access padlockePrints

Nonlinear 2.5D human face reconstruction from a single RGB image

Lookup NU author(s): Peng Liu, Dr Wai Lok Woo, Professor Satnam Dlay

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.


Abstract

Traditional 2D based image processing and face recognition is only considering gray level information, which is the combination of illumination, and object shape. However, shape and illumination are two independent components. This paper focus on separating illumination factors from 2D images by using ICA (Independent component analysis) method. Only a single image captured under RGB light sources is used as the Input. A nonlinear statistical reflection model is developed through the nonlinear ICA algorithm to counter balance nonlinear distortions during image capturing process. The proposed algorithm has achieved excellent features of separating the illumination factors and yielded very high accuracy of 2.5D human face data recovery.


Publication metadata

Author(s): Liu P, Woo WL, Dlay SS

Editor(s): Erich Leitgeb; Wolfgang Kogler; Zabih Ghassemlooy

Publication type: Conference Proceedings (inc. Abstract)

Publication status: Published

Conference Name: 6th International Symposium on Communication Systems, Networks and Digital Signal Processing (CSNDSP 2008)

Year of Conference: 2008

Pages: 613-617

ISSN: 9781424418756

Publisher: University of Graz

Library holdings: Search Newcastle University Library for this item

ISBN: 1424418755


Share