Reconstruction of coarse depth estimates using hierarchical image partitions

Authors:
Camilo Dorea;Ricardo L. Queiroz
Affiliations:
Department of Computer Science, University of Brasilia, Brasilia, Brazil 70919-970;Department of Computer Science, University of Brasilia, Brasilia, Brazil 70919-970
Venue:
Multidimensional Systems and Signal Processing
Year:
2013

Citing 8
Cited 0

A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms

International Journal of Computer Vision
High-quality video view interpolation using a layered representation

ACM SIGGRAPH 2004 Papers
Depth map creation and image-based rendering for advanced 3DTV services providing interoperability and scalability

Image Communication
Reconstruction of 3D models from intensity images and partial depth

AAAI'04 Proceedings of the 19th national conference on Artifical intelligence
Trajectory tree as an object-oriented hierarchical representation for video

IEEE Transactions on Circuits and Systems for Video Technology
Edge and motion-adaptive median filtering for multi-view depth map enhancement

PCS'09 Proceedings of the 27th conference on Picture Coding Symposium
Image quality assessment: from error visibility to structural similarity

IEEE Transactions on Image Processing
Binary Partition Tree Analysis Based on Region Evolution and Its Application to Tree Simplification

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel depth map enhancement method which takes as inputs a single view and an associated coarse depth estimate. The depth estimate is reconstructed according to the region boundaries proposed within a set of image partitions of the associated view. Image partitions are determined through a color-based region merging procedure and organized within a hierarchy. Novel depth estimates are computed according to plane-fitting procedures using the proposed regions as support and selected coarse depth estimates. The use of partition hierarchies provides an efficient mechanism for evaluating plane-fitting results over various scales and, in this way, reduce the dependency of the results on the quality of a single color-based segmentation. Experimental results on publicly available test sequences illustrate the potential of the algorithm in significantly improving low quality depth estimates.