A distortion evaluation framework in 3D video view synthesis

Purica, Andrei; Cagnazzo, Marco; Pesquet‐Popescu, Béatrice; Dufaux, Frédéric; Ionescu, Bogdan

doi:10.1109/ic3d.2015.7391827

Cited by 2 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because depth maps are subjected to distortions from the acquiring device or transmission systems, the synthesized image can be subjected to geometrical distortion of foreground objects and also poor reproduction of complex textures. As noted in other studies [26] [25] [27], traditional metrics such as PSNR or SSIM may not be the best way to asses the quality of synthesized images. This behavior can be explained by the strong correlation between scene geometry and position of highly distorted areas.…”

Section: Towards a Region Of Interest Evaluationmentioning

confidence: 97%

“…Purica et al [26] study the difference between encoding and synthesis artifacts and propose a ROI based SSIM by separating between encoding errors coming from the reference view and distortion caused by the DIBR warping process. In this paper we extend the ideas presented in [26] and propose a new ROI generation technique for SSIM evaluation of synthesized videos. Next, we perform a study of the results using a subjective evaluation database in order to validate our assumptions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Using region-of-interest for quality evaluation of DIBR-based view synthesis methods

Purica

Valenzise

Pesquet‐Popescu

et al. 2016

2016 Eighth International Conference on Quality of Multimedia Experience (QoMEX)

Self Cite

View full text Add to dashboard Cite

Abstract-As 3D media became more and more popular over the last years, new technologies are needed in the transmission, compression and creation of 3D content. One of the most commonly used techniques for aiding with the compression and creation of 3D content is known as view synthesis. The most effective class of view synthesis algorithms are using DepthImage-Based-Rendering techniques, which use explicit scene geometry to render new views. However, these methods may produce geometrical distortions and localized artifacts which are difficult to evaluate as they are inherently different from encoding errors and they are perceived differently by human subjects. In this paper, we propose a region-of-interest evaluation technique for view synthesis based on DIBR methods. Based on the assumption that certain areas determined by the geometrical properties of the scene are prone to distortions, we select a ROI by analyzing the multiple DIBR methods together with the ground truth. The approach is tested using a subjective evaluation view synthesis database and show that our method improves the SSIM correlation with subjective scores We also test another similar method and traditional metrics.

show abstract

Section: Towards a Region Of Interest Evaluationmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%