Depth sensing with disparity space images and three-dimensional recursive search
We present a coarse-to-fine stereo matching optimization applicable to methods utilizing the Disparity Space Image (DSI) structure. With the Three-dimensional Recursive Search algorithm (3DRS), a coarse disparity seed is obtained first, with minimal computational effort. The coarse disparity seed is...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2018-04-01
|
| Series: | Automatika |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/00051144.2018.1503137 |
| _version_ | 1811296031223054336 |
|---|---|
| author | Miroslav Rožić Tomislav Pribanić |
| author_facet | Miroslav Rožić Tomislav Pribanić |
| author_sort | Miroslav Rožić |
| collection | DOAJ |
| description | We present a coarse-to-fine stereo matching optimization applicable to methods utilizing the Disparity Space Image (DSI) structure. With the Three-dimensional Recursive Search algorithm (3DRS), a coarse disparity seed is obtained first, with minimal computational effort. The coarse disparity seed is then used as a guidance to locally compute the DSI disparity space with a reduced number of disparity hypotheses, yielding significantly shorter execution times for the disparity computation. The method performance was measured on the well-known Dynamic Programming (DP) DSI-based method and the images from the Middlebury set. The DP method with the DSI optimization applied maintains or improves the overall level of disparity map accuracy while delivering a near sevenfold speed-up of execution in comparison to DP alone. We furthermore show that the optimized method's performance does not depend on the expected input disparity range, which is commonly restricted, or expected to be defined upfront, for DSI-based stereo matching methods. |
| first_indexed | 2024-04-13T05:42:24Z |
| format | Article |
| id | doaj.art-9309b8fcfd994aa89045fdf1f9a5d33b |
| institution | Directory Open Access Journal |
| issn | 0005-1144 1848-3380 |
| language | English |
| last_indexed | 2024-04-13T05:42:24Z |
| publishDate | 2018-04-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Automatika |
| spelling | doaj.art-9309b8fcfd994aa89045fdf1f9a5d33b2022-12-22T03:00:04ZengTaylor & Francis GroupAutomatika0005-11441848-33802018-04-0159213114210.1080/00051144.2018.15031371503137Depth sensing with disparity space images and three-dimensional recursive searchMiroslav Rožić0Tomislav Pribanić1Zagrebačka Banka d.d.University of ZagrebWe present a coarse-to-fine stereo matching optimization applicable to methods utilizing the Disparity Space Image (DSI) structure. With the Three-dimensional Recursive Search algorithm (3DRS), a coarse disparity seed is obtained first, with minimal computational effort. The coarse disparity seed is then used as a guidance to locally compute the DSI disparity space with a reduced number of disparity hypotheses, yielding significantly shorter execution times for the disparity computation. The method performance was measured on the well-known Dynamic Programming (DP) DSI-based method and the images from the Middlebury set. The DP method with the DSI optimization applied maintains or improves the overall level of disparity map accuracy while delivering a near sevenfold speed-up of execution in comparison to DP alone. We furthermore show that the optimized method's performance does not depend on the expected input disparity range, which is commonly restricted, or expected to be defined upfront, for DSI-based stereo matching methods.http://dx.doi.org/10.1080/00051144.2018.1503137Stereo visiondisparity computationdynamic programmingDSI3DRScensus transform |
| spellingShingle | Miroslav Rožić Tomislav Pribanić Depth sensing with disparity space images and three-dimensional recursive search Automatika Stereo vision disparity computation dynamic programming DSI 3DRS census transform |
| title | Depth sensing with disparity space images and three-dimensional recursive search |
| title_full | Depth sensing with disparity space images and three-dimensional recursive search |
| title_fullStr | Depth sensing with disparity space images and three-dimensional recursive search |
| title_full_unstemmed | Depth sensing with disparity space images and three-dimensional recursive search |
| title_short | Depth sensing with disparity space images and three-dimensional recursive search |
| title_sort | depth sensing with disparity space images and three dimensional recursive search |
| topic | Stereo vision disparity computation dynamic programming DSI 3DRS census transform |
| url | http://dx.doi.org/10.1080/00051144.2018.1503137 |
| work_keys_str_mv | AT miroslavrozic depthsensingwithdisparityspaceimagesandthreedimensionalrecursivesearch AT tomislavpribanic depthsensingwithdisparityspaceimagesandthreedimensionalrecursivesearch |