A Dynamic Programming Approach to Reconstructing Building Interiors
A number of recent papers have investigated reconstruction under Manhattan world assumption, in which surfaces in the world are assumed to be aligned with one of three dominant directions [1,2,3,4]. In this paper we present a dynamic programming solution to the reconstruction problem for "indoo...
| Main Authors: | , , , |
|---|---|
| Format: | Conference item |
| Published: |
2010
|
| _version_ | 1797069135555854336 |
|---|---|
| author | Flint, A Mei, C Murray, D Reid, I |
| author_facet | Flint, A Mei, C Murray, D Reid, I |
| author_sort | Flint, A |
| collection | OXFORD |
| description | A number of recent papers have investigated reconstruction under Manhattan world assumption, in which surfaces in the world are assumed to be aligned with one of three dominant directions [1,2,3,4]. In this paper we present a dynamic programming solution to the reconstruction problem for "indoor" Manhattan worlds (a sub-class of Manhattan worlds). Our algorithm deterministically finds the global optimum and exhibits computational complexity linear in both model complexity and image size. This is an important improvement over previous methods that were either approximate [3] or exponential in model complexity [4]. We present results for a new dataset containing several hundred manually annotated images, which are released in conjunction with this paper. © 2010 Springer-Verlag. |
| first_indexed | 2024-03-06T22:20:02Z |
| format | Conference item |
| id | oxford-uuid:54b54db1-7a35-48a7-98c6-1f1d83dd0f92 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T22:20:02Z |
| publishDate | 2010 |
| record_format | dspace |
| spelling | oxford-uuid:54b54db1-7a35-48a7-98c6-1f1d83dd0f922022-03-26T16:39:37ZA Dynamic Programming Approach to Reconstructing Building InteriorsConference itemhttp://purl.org/coar/resource_type/c_5794uuid:54b54db1-7a35-48a7-98c6-1f1d83dd0f92Symplectic Elements at Oxford2010Flint, AMei, CMurray, DReid, IA number of recent papers have investigated reconstruction under Manhattan world assumption, in which surfaces in the world are assumed to be aligned with one of three dominant directions [1,2,3,4]. In this paper we present a dynamic programming solution to the reconstruction problem for "indoor" Manhattan worlds (a sub-class of Manhattan worlds). Our algorithm deterministically finds the global optimum and exhibits computational complexity linear in both model complexity and image size. This is an important improvement over previous methods that were either approximate [3] or exponential in model complexity [4]. We present results for a new dataset containing several hundred manually annotated images, which are released in conjunction with this paper. © 2010 Springer-Verlag. |
| spellingShingle | Flint, A Mei, C Murray, D Reid, I A Dynamic Programming Approach to Reconstructing Building Interiors |
| title | A Dynamic Programming Approach to Reconstructing Building Interiors |
| title_full | A Dynamic Programming Approach to Reconstructing Building Interiors |
| title_fullStr | A Dynamic Programming Approach to Reconstructing Building Interiors |
| title_full_unstemmed | A Dynamic Programming Approach to Reconstructing Building Interiors |
| title_short | A Dynamic Programming Approach to Reconstructing Building Interiors |
| title_sort | dynamic programming approach to reconstructing building interiors |
| work_keys_str_mv | AT flinta adynamicprogrammingapproachtoreconstructingbuildinginteriors AT meic adynamicprogrammingapproachtoreconstructingbuildinginteriors AT murrayd adynamicprogrammingapproachtoreconstructingbuildinginteriors AT reidi adynamicprogrammingapproachtoreconstructingbuildinginteriors AT flinta dynamicprogrammingapproachtoreconstructingbuildinginteriors AT meic dynamicprogrammingapproachtoreconstructingbuildinginteriors AT murrayd dynamicprogrammingapproachtoreconstructingbuildinginteriors AT reidi dynamicprogrammingapproachtoreconstructingbuildinginteriors |