by T Vaudrey, R Klette
Abstract:
In past studies, illumination effects have been proven to cause the most common problems in correspondence algorithms. In this paper, we conduct a study identifying that the residual images (i.e., differences between images and their smoothed versions) contain the important information in an image. We go on to show that this approach removes illumination artifacts between corresponding pairs of images (i.e., optical flow and stereo) using a mixture of synthetic and real-life images. © 2009 Springer Berlin Heidelberg.
Reference:
Residual images remove illumination artifacts! (T Vaudrey, R Klette), In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), volume 5748 LNCS, 2009.
Bibtex Entry:
@inproceedings{vaudrey2009residualartifacts!,
author = "Vaudrey, T and Klette, R",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "472--481",
title = "Residual images remove illumination artifacts!",
volume = "5748 LNCS",
year = "2009",
abstract = "In past studies, illumination effects have been proven to cause the most common problems in correspondence algorithms. In this paper, we conduct a study identifying that the residual images (i.e., differences between images and their smoothed versions) contain the important information in an image. We go on to show that this approach removes illumination artifacts between corresponding pairs of images (i.e., optical flow and stereo) using a mixture of synthetic and real-life images. © 2009 Springer Berlin Heidelberg.",
doi = "10.1007/978-3-642-03798-6_48",
isbn = "3642037976",
isbn = "9783642037979",
issn = "0302-9743",
eissn = "1611-3349",
language = "eng",
}