Fast correlation-based stereo matching with the reduction of systematic errors
Pattern Recognition Letters
View synthesis by the parallel use of GPU and CPU
Image and Vision Computing
Depth compositing for augmented reality
ACM SIGGRAPH 2008 posters
Probabilistic Scene Analysis for Robust Stereo Correspondence
ICIAR '09 Proceedings of the 6th International Conference on Image Analysis and Recognition
Depth Imaging by Combining Time-of-Flight and On-Demand Stereo
Dyn3D '09 Proceedings of the DAGM 2009 Workshop on Dynamic 3D Imaging
A novel optical see-through head-mounted display with occlusion and intensity matching support
Edutainment'07 Proceedings of the 2nd international conference on Technologies for e-learning and digital entertainment
| Hi-index | 0.00 |
This work presents a technique for computing dense disparitymaps from a binocular stereo camera system. Themethods are applied in an Augmented Reality setting forcombining real and virtual worlds with proper occlusions.The proposed stereo correspondence technique is based onarea matching and facilitates an efficient strategy by usingthe concept of a three-dimensional similarity accumulator,whereby occlusions are detected and object boundaries areextracted correctly. The main contribution of this paper isthe way we fill the accumulator using absolute differencesof images and computing a mean filter on these differenceimages. This is where the main advantages of the accumulatorapproach can be exploited, since all entries canbe computed in parallel and thus extremely efficient. Additionally,we perform an asymmetric correction step and apost-processing of the disparity maps that maintains objectedges.