Publication View

3D object-based depth estimation from stereoscopic image sequences (1995)

Abstract
A 3D object--based depth estimation approach from stereoscopic image sequences is presented in this paper that introduces a memory for depth information from preceding image pairs. This approach is used in the EU--RACE project DISTIMA for applications like remote handling, quality control and TV post processing. Compared to depth estimation approaches that compute depth maps from consecutive stereoscopic image pairs independently by a triangulation of corresponding image points, the temporal coherence and accuracy of depth estimates is increased. The remaining depth estimation error variance is reduced and as a result of temporal filtering the effect of single outliers is reduced as well. 1.

Publication details
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=?doi=10.1.1.30.7283
Source ftp://ftp.tnt.uni-hannover.de/pub/papers/1995/IWS3DI95_LF.ps.gz
Contributors CiteSeerX
Repository CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Type text
Language English
Relation 10.1.1.45.6056, 10.1.1.112.8374, 10.1.1.45.4255, 10.1.1.72.5799, 10.1.1.21.418, 10.1.1.56.1246, 10.1.1.21.3439, 10.1.1.45.1644, 10.1.1.57.6054