Loading...
Please wait, while we are loading the content...
Similar Documents
Efficient dense depth estimation from dense multiperspective panoramas (2001)
| Content Provider | CiteSeerX |
|---|---|
| Author | Tang, Chi-Keung Li, Yin |
| Description | in Proc. International Conference on Computer Vision (ICCV |
| Abstract | In this puper we study how to compute U dense depth map with punorumic jield oj ' view (e.g., 360 degrees) from multi-perspective punommus. A dense sequence of multiperspective punorumus is used fiw better uccurucy and reduced ambiguity by tuking udvuntuge oj ' signijicunt dutu redunduncy. To speed up the reconstruction, we derive un upproximute epipolar plune imuge thut is ussociuted with the planar sweeping cameru setup. und use one-dimensional window jor efficient mutching. To uddress the uperture problem introduced by one-dimensional window matching, we keep U set oj ' possible depth cundidutes,from mutching scores. These cundidutes ure then pussed to U novel rwo-pu.s.s tensor voting scheme to select the optimal depth. By propuguting the continuity und uniqueness construints non-iterutively in the voting process, our method produces high-quulity reconstruction results even when signiJlcunt occlusion is present. Experiments on chullenging sjnthetic und real scenes demonstrutr the eflectiveness und eflcucy oj'our method. 1 |
| File Format | |
| Publisher Date | 2001-01-01 |
| Access Restriction | Open |
| Subject Keyword | One-dimensional Window Matching High-quulity Reconstruction Result Optimal Depth Signijlcunt Occlusion Signijicunt Dutu Redunduncy Uniqueness Construints Reduced Ambiguity Punorumic Jield Oj Sjnthetic Und Real Scene Cameru Setup Uperture Problem Possible Depth Cundidutes Voting Process Multi-perspective Punommus Efficient Dense Depth Estimation One-dimensional Window Jor Efficient Mutching Udvuntuge Oj Novel Rwo-pu Dense Depth Map Dense Sequence Multiperspective Punorumus |
| Content Type | Text |
| Resource Type | Conference Proceedings |
