Estimating Illumination Direction of 3D Surface Texture Based on Active Basis and Mojette Transform

This paper describes an approach that, given an image of a rough surface texture, can estimate the direction from which the surface was lit. Unlike previous work, we require neither surface isotropy nor that the texture is included in the training set. The approach is based on Active Basis and the Mojette transform. The Mojette transform is used to estimate the orientation of the texture; it does so by finding which training samples have similar orientation features. The Active Basis model is then learned from training images by the shared pursuit algorithm. Next, the base histograms of the test image and textures with similar orientation features in the training set are compared so that the illumination directions can be estimated by minimizing their correlation coefficients.

[1]  Junyu Dong,et al.  Directionality measurement and illumination estimation of 3D surface textures by using mojette transform , 2008, 2008 19th International Conference on Pattern Recognition.

[2]  Andrew Zisserman,et al.  Estimating illumination direction from textured images , 2004, CVPR 2004.

[3]  V. Vijaya Kumar,et al.  Texture Classification by Simple Patterns on Edge Direction Movements , 2007 .

[4]  G. Vaidyanathan,et al.  Texture direction analysis using edge counts , 1989, Proceedings. IEEE Energy and Information Technologies in the Southeast'.

[5]  Hamid Soltanian-Zadeh,et al.  Radon transform orientation estimation for rotation invariant texture analysis , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[7]  Florent Autrusseau,et al.  Redundant image representation via multi-scale digital Radon projections , 2008, 2008 15th IEEE International Conference on Image Processing.

[8]  Florent Autrusseau,et al.  Lossless image compression via predictive coding of discrete Radon projections , 2008, Signal Process. Image Commun..

[9]  Song-Chun Zhu,et al.  Deformable Template As Active Basis , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[10]  Stefan M. Karlsson,et al.  Illuminance flow over anisotropic surfaces. , 2008, Journal of the Optical Society of America. A, Optics, image science, and vision.

[11]  Deng-Yi Zhang,et al.  Object tracking by using active basis model , 2009, International Symposium on Multispectral Image Processing and Pattern Recognition.

[12]  Maria Petrou,et al.  Estimating Lighting Direction and Classifying Textures , 2002, BMVC.

[13]  Song-Chun Zhu,et al.  Learning Active Basis Model for Object Detection and Recognition , 2010, International Journal of Computer Vision.

[14]  Jean-Pierre V. Guédon,et al.  Spline Mojette transform. Applications in tomography and communications , 2002, 2002 11th European Signal Processing Conference.

[15]  Jan-Olof Eklundh,et al.  Automatic estimation of the projected light source direction , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[16]  Maria Petrou,et al.  Classifying Surface Texture while Simultaneously Estimating Illumination Direction , 2005, International Journal of Computer Vision.

[17]  Rama Chellappa,et al.  Estimation of Illuminant Direction, Albedo, and Shape from Shading , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  J. Koenderink,et al.  Irradiation direction from texture. , 2003, Journal of the Optical Society of America. A, Optics, image science, and vision.