Deriving intrinsic images from image sequences

Y. Weiss*

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

534 Scopus citations

Abstract

Intrinsic images are a useful midlevel description of scenes proposed by Barrow and Tenenbaum [1]. An image is decomposed into two images: a reflectance image and an illumination image. Finding such a decomposition remains a difficult problem in computer vision. Here we focus on a slightly easier problem: given a sequence of T images where the reflectance is constant and the illumination changes, can we recover T illumination images and a single reflectance image? We show that this problem is still illposed and suggest approaching it as a maximum-likelihood estimation problem. Following recent work on the statistics of natural images, we use a prior that assumes that illumination images will give rise to sparse filter outputs. We show that this leads to a simple, novel algorithm for recovering reflectance images. We illustrate the algorithm's performance on real and synthetic image sequences.

Original languageAmerican English
Pages68-75
Number of pages8
StatePublished - 2001
Externally publishedYes
Event8th International Conference on Computer Vision - Vancouver, BC, United States
Duration: 9 Jul 200112 Jul 2001

Conference

Conference8th International Conference on Computer Vision
Country/TerritoryUnited States
CityVancouver, BC
Period9/07/0112/07/01

Fingerprint

Dive into the research topics of 'Deriving intrinsic images from image sequences'. Together they form a unique fingerprint.

Cite this