The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
L'estimation de la profondeur (disparité) à partir d'un champ lumineux (un ensemble d'images denses à vues multiples) suscite actuellement beaucoup d'intérêt dans la recherche. Cet article se concentre sur la façon de gérer un champ lumineux bruyant pour l'estimation de la disparité, car s'il est laissé tel quel, le bruit détériore la précision des cartes de disparité estimées. Plusieurs chercheurs ont travaillé sur ce problème, par exemple en introduisant des signaux de disparité robustes au bruit. Cependant, il n’est pas facile de trouver un compromis entre précision et vitesse de calcul. Pour résoudre ce compromis, nous avons intégré un schéma de débruitage rapide dans un cadre d'estimation de disparité rapide qui fonctionne dans le domaine de l'image plan épipolaire (EPI). Plus précisément, nous avons constaté qu’un simple filtre incliné 1D est très efficace pour réduire le bruit tout en préservant la structure sous-jacente d’un EPI. De plus, ce filtrage simple ne nécessite pas de configurations de paramètres élaborées en fonction du niveau de bruit cible. Les résultats expérimentaux incluant des entrées du monde réel montrent que notre méthode peut atteindre une bonne précision avec beaucoup moins de temps de calcul par rapport à certaines méthodes de pointe.
Gou HOUBEN
Nagoya University
Shu FUJITA
Nagoya University
Keita TAKAHASHI
Nagoya University
Toshiaki FUJII
Nagoya University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Gou HOUBEN, Shu FUJITA, Keita TAKAHASHI, Toshiaki FUJII, "Fast and Robust Disparity Estimation from Noisy Light Fields Using 1-D Slanted Filters" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 11, pp. 2101-2109, November 2019, doi: 10.1587/transinf.2019PCP0003.
Abstract: Depth (disparity) estimation from a light field (a set of dense multi-view images) is currently attracting much research interest. This paper focuses on how to handle a noisy light field for disparity estimation, because if left as it is, the noise deteriorates the accuracy of estimated disparity maps. Several researchers have worked on this problem, e.g., by introducing disparity cues that are robust to noise. However, it is not easy to break the trade-off between the accuracy and computational speed. To tackle this trade-off, we have integrated a fast denoising scheme in a fast disparity estimation framework that works in the epipolar plane image (EPI) domain. Specifically, we found that a simple 1-D slanted filter is very effective for reducing noise while preserving the underlying structure in an EPI. Moreover, this simple filtering does not require elaborate parameter configurations in accordance with the target noise level. Experimental results including real-world inputs show that our method can achieve good accuracy with much less computational time compared to some state-of-the-art methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019PCP0003/_p
Copier
@ARTICLE{e102-d_11_2101,
author={Gou HOUBEN, Shu FUJITA, Keita TAKAHASHI, Toshiaki FUJII, },
journal={IEICE TRANSACTIONS on Information},
title={Fast and Robust Disparity Estimation from Noisy Light Fields Using 1-D Slanted Filters},
year={2019},
volume={E102-D},
number={11},
pages={2101-2109},
abstract={Depth (disparity) estimation from a light field (a set of dense multi-view images) is currently attracting much research interest. This paper focuses on how to handle a noisy light field for disparity estimation, because if left as it is, the noise deteriorates the accuracy of estimated disparity maps. Several researchers have worked on this problem, e.g., by introducing disparity cues that are robust to noise. However, it is not easy to break the trade-off between the accuracy and computational speed. To tackle this trade-off, we have integrated a fast denoising scheme in a fast disparity estimation framework that works in the epipolar plane image (EPI) domain. Specifically, we found that a simple 1-D slanted filter is very effective for reducing noise while preserving the underlying structure in an EPI. Moreover, this simple filtering does not require elaborate parameter configurations in accordance with the target noise level. Experimental results including real-world inputs show that our method can achieve good accuracy with much less computational time compared to some state-of-the-art methods.},
keywords={},
doi={10.1587/transinf.2019PCP0003},
ISSN={1745-1361},
month={November},}
Copier
TY - JOUR
TI - Fast and Robust Disparity Estimation from Noisy Light Fields Using 1-D Slanted Filters
T2 - IEICE TRANSACTIONS on Information
SP - 2101
EP - 2109
AU - Gou HOUBEN
AU - Shu FUJITA
AU - Keita TAKAHASHI
AU - Toshiaki FUJII
PY - 2019
DO - 10.1587/transinf.2019PCP0003
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2019
AB - Depth (disparity) estimation from a light field (a set of dense multi-view images) is currently attracting much research interest. This paper focuses on how to handle a noisy light field for disparity estimation, because if left as it is, the noise deteriorates the accuracy of estimated disparity maps. Several researchers have worked on this problem, e.g., by introducing disparity cues that are robust to noise. However, it is not easy to break the trade-off between the accuracy and computational speed. To tackle this trade-off, we have integrated a fast denoising scheme in a fast disparity estimation framework that works in the epipolar plane image (EPI) domain. Specifically, we found that a simple 1-D slanted filter is very effective for reducing noise while preserving the underlying structure in an EPI. Moreover, this simple filtering does not require elaborate parameter configurations in accordance with the target noise level. Experimental results including real-world inputs show that our method can achieve good accuracy with much less computational time compared to some state-of-the-art methods.
ER -