Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Benchmarking of several disparity estimation algorithms for light field processing

 
: Zakeri, Faezeh Sadat; Bätz, Michel; Jaschke, Tobias; Keinert, Joachim; Chuchvara, Alexandra

:
Fulltext urn:nbn:de:0011-n-5524878 (1.0 MByte PDF)
MD5 Fingerprint: 2d0c779062a2c31cad6ae9e923b5d400
Copyright Society of Photo-Optical Instrumentation Engineers. One print or electronic copy may be made for personal use only. Systematic reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited.
Created on: 31.7.2019


Cudel, Christophe ; Society of Photo-Optical Instrumentation Engineers -SPIE-, Bellingham/Wash.:
14th International Conference on Quality Control by Artificial Vision 2019 : 15-17 May 2019, Mulhouse, France
Bellingham, WA: SPIE, 2019 (Proceedings of SPIE 11172)
ISBN: 978-1-5106-3053-6
ISBN: 978-1-5106-3054-3
Paper 111721C, 10 pp.
International Conference on Quality Control by Artificial Vision <14, 2019, Mulhouse>
European Commission EC
H2020; 676401; ETN-FPI
European Training Network on Full Parallax Imaging
English
Conference Paper, Electronic Publication
Fraunhofer IIS ()
3D image processing; reconstruction algorithm; image quality; lightfield

Abstract
A number of high-quality depth imaged-based rendering (DIBR) pipelines have been developed to reconstruct a 3D scene from several images taken from known camera viewpoints. Due to the specific limitations of each technique, their output is prone to artifacts. Therefore, the quality cannot be ensured. To improve the quality of the most critical and challenging image areas, an exhaustive comparison is required. In this paper, we consider three questions of benchmarking the quality performance of eight DIBR techniques on light fields: First, how does the density of original input views affect the quality of the rendered novel views? Second, how does disparity range between adjacent input views impact the quality? Third, how does each technique behave for different object properties? We compared and evaluated the results visually as well as quantitatively (PSNR, SSIM, AD, and VDP2). The results show some techniques outperform others in different disparity ranges. The results also indicate using more views not necessarily results in visually higher quality for all critical image areas. Finally, we have shown a comparison for different scene’s complexity such as non-Lambertian objects.

: http://publica.fraunhofer.de/documents/N-552487.html