Consistent Depth Maps Recovery from a Video Sequence
Authors:
Top Cited Papers
- 6 March 2009
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 31 (6) , 974-988
Abstract
This paper presents a novel method for recovering consistent depth maps from a video sequence. We propose a bundle optimization framework to address the major difficulties in stereo reconstruction, such as dealing with image noise, occlusions, and outliers. Different from the typical multi-view stereo methods, our approach not only imposes the photo-consistency constraint, but also explicitly associates the geometric coherence with multiple frames in a statistical way. It thus can naturally maintain the temporal coherence of the recovered dense depth maps without over-smoothing. To make the inference tractable, we introduce an iterative optimization scheme by first initializing the disparity maps using a segmentation prior and then refining the disparities by means of bundle optimization. Instead of defining the visibility parameters, our method implicitly models the reconstruction noise as well as the probabilistic visibility. After bundle optimization, we introduce an efficient space-time fusion algorithm to further reduce the reconstruction noise. Our automatic depth recovery is evaluated using a variety of challenging video examples.Keywords
This publication has 50 references indexed in Scilit:
- Global stereo reconstruction under second order smoothness priors2008
- Stereo reconstruction with mixed pixels using adaptive over-segmentation2008
- TransforMesh : A Topology-Adaptive Mesh-Based Approach to Surface Evolution2007
- A Globally Optimal Algorithm for Robust TV-L1 Range Image Integration2007
- Multi-View Stereo for Community Photo Collections2007
- Stereo Matching with Color-Weighted Correlation, Hierachical Belief Propagation and Occlusion Handling2006
- Multi-View Stereo via Volumetric Graph-Cuts2005
- Comparison of graph cuts with belief propagation for stereo, using identical MRF parameters2003
- A global matching framework for stereo computation2002
- Multilevel computational processes for visual surface reconstructionComputer Vision, Graphics, and Image Processing, 1983