Median Filter Based Adaptive Compensation Method for Depth Map Pre-Processing
Asha Jayachandran1, Preetha V.H2

1Asha Jayachandran, PG Student, Department of Electronics and Communication Engineering, SCT College of Engineering, Pappanamcode, (Kerala), India.
2Preetha V.H, Assistant Professor, Department of Electronics and Communication Engineering, SCT College of Engineering, Pappanamcode, (Kerala), India.

Manuscript received on 13 June 2016 | Revised Manuscript received on 20 June 2016 | Manuscript Published on 30 June 2016 | PP: 94-99 | Volume-5 Issue-5, June 2016 | Retrieval Number: E4618065516/16©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Depth Image Based Rendering (DIBR) is 2D to 3D conversion technology using color image and its corresponding depth image that is widely employed in applications like 3D TV, free view television etc. 3-D viewing is the next most happening technology. Since transmission of 3D video demands a lot of bandwidth, a new technology that renders virtual views using a color image and its corresponding depth image was proposed. If the depth map is incomplete, the virtual views generated will contain holes or disocclusions which affect the quality of 3D viewing. Since holes occur when the intensity in depth map changes significantly, smoothening methods were proposed reduce the number of holes. Since smoothening methods affect the edges and destroys the original information in the depth map, Adaptive Compensation method (ADC) which processes the image in different modes was proposed. Improved Adaptive Compensation method does not produce satisfactory results for images with large number of holes. Though an improvement in PSNR and SSIM improvement is observed, the number of holes in the warped image is increased. A median filtering is incorporated in Adaptive Compensation method to reduce the number of holes. The experimental results indicate an improvement in PSNR and SSIM as well as a reduction in number of holes.
Keywords: Depth Image Based Rendering, 3D TV, Inpainting, Adaptive Compensation Method, Disocclusion, Median Filter, Holes, Virtual views.

Scope of the Article: Image analysis and Processing