Loughborough University
Leicestershire, UK
LE11 3TU
+44 (0)1509 263171
Loughborough University

Loughborough University Institutional Repository

Please use this identifier to cite or link to this item: https://dspace.lboro.ac.uk/2134/22793

Title: Edge and motion-adaptive median filtering for multi-view depth map enhancement
Authors: Ekmekcioglu, Erhan
Velisavljevic, Vladan
Worrall, Stewart T.
Keywords: Depth estimation
Free viewpoint video
Multiview coding
Multi-view depth map
Issue Date: 2009
Publisher: © IEEE
Citation: EKMEKCIOGLU, E., VELISAVLJEVIC, V. and WORRALL, S., 2009. Edge and motion-adaptive median filtering for multi-view depth map enhancement. IN: Proceedings of 2009 IEEE Picture Coding Symposium (PCS 2009), Chicago, United States, 6-8 May 2009, doi: 10.1109/PCS.2009.5167415.
Abstract: We present a novel multi-view depth map enhancement method deployed as a post-processing of initially estimated depth maps, which are incoherent in the temporal and inter-view dimensions. The proposed method is based on edge and motion-adaptive median filtering and allows for an improved quality of virtual view synthesis. To enforce the spatial, temporal and inter-view coherence in the multiview depth maps, the median filtering is applied to 4-dimensional windows that consist of the spatially neighbor depth map values taken at different viewpoints and time instants. These windows have locally adaptive shapes in a presence of edges or motion to preserve sharpness and realistic rendering. We show that our enhancement method leads to a reduction of a coding bit-rate required for representation of the depth maps and also to a gain in the quality of synthesized views at an arbitrary virtual viewpoint. At the same time, the method carries a low additional computational complexity.
Description: Closed access.
Sponsor: This work was in part developed within VISNET II, a European Network of Excellence (http://www.visnetnoe.org), funded under the European Commission IST FP6 programme.
Version: Published
DOI: 10.1109/PCS.2009.5167415
URI: https://dspace.lboro.ac.uk/2134/22793
Publisher Link: http://dx.doi.org/10.1109/PCS.2009.5167415
ISBN: 1424445930
Appears in Collections:Closed Access (Loughborough University London)

Files associated with this item:

File Description SizeFormat
05167415.pdfPublished version303.79 kBAdobe PDFView/Open


SFX Query

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.