Modelling Needle-Map Consistency with Novel Constraints

Authors:
Philip L. Worthington;Edwin R. Hancock
Affiliations:
-;-
Venue:
CAIP '99 Proceedings of the 8th International Conference on Computer Analysis of Images and Patterns
Year:
1999

Citing 7
Cited 0

Improved methods of estimating shape from shading using the light source coordinate system

Artificial Intelligence
Computational vision and regularization theory

Nature
Shape from shading

Shape from shading
Height and gradient from shading

International Journal of Computer Vision
Estimation of Illuminant Direction, Albedo, and Shape from Shading

IEEE Transactions on Pattern Analysis and Machine Intelligence
Surface shape and curvature scales

Image and Vision Computing
Shape and source from shading

IJCAI'85 Proceedings of the 9th international joint conference on Artificial intelligence - Volume 2

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses two fundamental criticisms of existing global shape-from-shading techniques. Firstly, we introduce a new update framework which ensures data-closeness at all stages whilst iteratively recovering the needle-map by treating the image irradiance equation as a hard constraint. Secondly, we apply the apparatus of robust statistics and novel constraints on needle-map smoothness to accomodate surface orientation discontinuities in a principled fashion.