Illuminance Flow Estimation by Regression

Journal article (2010)

Authors

S.M. Karlsson

S.C. Pont

J.J. Koenderink

A. Zisserman

Department

Industrial Design () (TU Delft)

DOI: https://doi.org/doi:10.1007/s11263-010-0353-7

Illuminance flow Surface 3D texture Histogram of oriented gradients Illuminant estimation

To reference this document use:

http://resolver.tudelft.nl/uuid:e6bf6341-d6e9-443f-bed8-8afd6c57a00c

More Info

expand_more

Published Date

09-06-2010

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Source:

International Journal of Computer Vision, 90 (3), 2010

Faculty

Industrial Design Engineering

Department

Industrial Design

Abstract

We investigate the estimation of illuminance flow using Histograms of Oriented Gradient features (HOGs). In a regression setting, we found for both ridge regression and support vector machines, that the optimal solution shows close resemblance to the gradient based structure tensor (also known as the second moment matrix). Theoretical results are presented showing in detail how the structure tensor and the HOGs are connected. This relation will benefit computer vision tasks such as affine invariant texture/object matching using HOGs. Several properties of HOGs are presented, among others, how many bins are required for a directionality measure, and how to estimate HOGs through spatial averaging that requires no binning.

Files

Karlsson.pdf

(pdf | 0.518 Mb)