A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations
International audience Given a pair of stereo images, the spatial coordinates of a scene point can be derived from its projections onto the two considered image planes. Finding the correspondences between such projections however remains the main difficulty of the depth estimation problem: the match...
Main Authors: | , , |
---|---|
Other Authors: | , , , , , |
Format: | Conference Object |
Language: | English |
Published: |
HAL CCSD
2015
|
Subjects: | |
Online Access: | https://hal.science/hal-01158707 https://doi.org/10.1007/978-3-319-18720-4_11 |
id |
ftminesparistech:oai:HAL:hal-01158707v1 |
---|---|
record_format |
openpolar |
spelling |
ftminesparistech:oai:HAL:hal-01158707v1 2024-05-19T07:42:50+00:00 A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations Bricola, Jean-Charles Bilodeau, Michel Beucher, Serge Centre de Morphologie Mathématique (CMM) Mines Paris - PSL (École nationale supérieure des mines de Paris) Université Paris Sciences et Lettres (PSL)-Université Paris Sciences et Lettres (PSL) ARMINES / CMM Jón Atli Benediktsson European Project: 296104,EC:FP7:SP1-JTI,ENIAC-2011-1,PANORAMA(2012) Reykjavík, Iceland 2015-05-27 https://hal.science/hal-01158707 https://doi.org/10.1007/978-3-319-18720-4_11 en eng HAL CCSD info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-319-18720-4_11 info:eu-repo/grantAgreement/EC/FP7/296104/EU/Ultra Wide Context Aware Imaging/PANORAMA hal-01158707 https://hal.science/hal-01158707 doi:10.1007/978-3-319-18720-4_11 Mathematical Morphology and Its Applications to Signal and Image Processing.12th International Symposium, ISMM 2015, Reykjavik, Iceland, May 27-29, 2015. Proceedings International Symposium on Mathematical Morphology 2015 https://hal.science/hal-01158707 International Symposium on Mathematical Morphology 2015, Jón Atli Benediktsson, May 2015, Reykjavík, Iceland. ⟨10.1007/978-3-319-18720-4_11⟩ http://link.springer.com/chapter/10.1007/978-3-319-18720-4_11 watershed segmentation hierarchies disparity estimation joint stereo segmentation non-ideal stereo imagery [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] info:eu-repo/semantics/conferenceObject Conference papers 2015 ftminesparistech https://doi.org/10.1007/978-3-319-18720-4_11 2024-04-25T00:50:46Z International audience Given a pair of stereo images, the spatial coordinates of a scene point can be derived from its projections onto the two considered image planes. Finding the correspondences between such projections however remains the main difficulty of the depth estimation problem: the matching of points across homogeneous regions is ambiguous and occluded points cannot be matched as their projections do not exist in one of the image planes.Instead of searching for dense point correspondences, this article proposes an approach to the estimation of depth map which is based on the matching of regions. The matchings are performed at two segmentation levels obtained by morphological criteria which ensure the existence of an hierarchy between the coarse and fine partitions. The hierarchy is then exploited in order to compute fine regional disparity maps which are accurate and free from noisy measurements.We finally show how this method fits to different sorts of stereo images: those which are highly textured, taken under constant illumination such as Middlebury and those which relevant information resides in the contours only. Conference Object Iceland Reykjavík Reykjavík MINES ParisTech: Open Archive (HAL) 122 133 |
institution |
Open Polar |
collection |
MINES ParisTech: Open Archive (HAL) |
op_collection_id |
ftminesparistech |
language |
English |
topic |
watershed segmentation hierarchies disparity estimation joint stereo segmentation non-ideal stereo imagery [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] |
spellingShingle |
watershed segmentation hierarchies disparity estimation joint stereo segmentation non-ideal stereo imagery [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] Bricola, Jean-Charles Bilodeau, Michel Beucher, Serge A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations |
topic_facet |
watershed segmentation hierarchies disparity estimation joint stereo segmentation non-ideal stereo imagery [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] |
description |
International audience Given a pair of stereo images, the spatial coordinates of a scene point can be derived from its projections onto the two considered image planes. Finding the correspondences between such projections however remains the main difficulty of the depth estimation problem: the matching of points across homogeneous regions is ambiguous and occluded points cannot be matched as their projections do not exist in one of the image planes.Instead of searching for dense point correspondences, this article proposes an approach to the estimation of depth map which is based on the matching of regions. The matchings are performed at two segmentation levels obtained by morphological criteria which ensure the existence of an hierarchy between the coarse and fine partitions. The hierarchy is then exploited in order to compute fine regional disparity maps which are accurate and free from noisy measurements.We finally show how this method fits to different sorts of stereo images: those which are highly textured, taken under constant illumination such as Middlebury and those which relevant information resides in the contours only. |
author2 |
Centre de Morphologie Mathématique (CMM) Mines Paris - PSL (École nationale supérieure des mines de Paris) Université Paris Sciences et Lettres (PSL)-Université Paris Sciences et Lettres (PSL) ARMINES / CMM Jón Atli Benediktsson European Project: 296104,EC:FP7:SP1-JTI,ENIAC-2011-1,PANORAMA(2012) |
format |
Conference Object |
author |
Bricola, Jean-Charles Bilodeau, Michel Beucher, Serge |
author_facet |
Bricola, Jean-Charles Bilodeau, Michel Beucher, Serge |
author_sort |
Bricola, Jean-Charles |
title |
A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations |
title_short |
A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations |
title_full |
A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations |
title_fullStr |
A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations |
title_full_unstemmed |
A Top-Down Approach to the Estimation of Depth Maps Driven by Morphological Segmentations |
title_sort |
top-down approach to the estimation of depth maps driven by morphological segmentations |
publisher |
HAL CCSD |
publishDate |
2015 |
url |
https://hal.science/hal-01158707 https://doi.org/10.1007/978-3-319-18720-4_11 |
op_coverage |
Reykjavík, Iceland |
genre |
Iceland Reykjavík Reykjavík |
genre_facet |
Iceland Reykjavík Reykjavík |
op_source |
Mathematical Morphology and Its Applications to Signal and Image Processing.12th International Symposium, ISMM 2015, Reykjavik, Iceland, May 27-29, 2015. Proceedings International Symposium on Mathematical Morphology 2015 https://hal.science/hal-01158707 International Symposium on Mathematical Morphology 2015, Jón Atli Benediktsson, May 2015, Reykjavík, Iceland. ⟨10.1007/978-3-319-18720-4_11⟩ http://link.springer.com/chapter/10.1007/978-3-319-18720-4_11 |
op_relation |
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-319-18720-4_11 info:eu-repo/grantAgreement/EC/FP7/296104/EU/Ultra Wide Context Aware Imaging/PANORAMA hal-01158707 https://hal.science/hal-01158707 doi:10.1007/978-3-319-18720-4_11 |
op_doi |
https://doi.org/10.1007/978-3-319-18720-4_11 |
container_start_page |
122 |
op_container_end_page |
133 |
_version_ |
1799482542840086528 |