Demand for 3D content increased over the last years and more attention was dedicated to methods used in 3D content creation and transmission. View synthesis methods introduce localized artifacts when creating new virtual views. Evaluating these methods require therefore a different approach, in order to identify and emphasize synthesis artifact prone areas, while diminishing the impact, of other types of artifacts, such as those produced by quantization during the video coding. In this paper, we propose an artifact evaluation technique which was shown to provide a better differentiation of synthesis methods, while additional information can be extrapolated from the results about the spatial localization of typical synthesis artifacts and their impact on overall image quality.
A distortion evaluation framework in 3D video view synthesis
Cagnazzo M.;
2016
Abstract
Demand for 3D content increased over the last years and more attention was dedicated to methods used in 3D content creation and transmission. View synthesis methods introduce localized artifacts when creating new virtual views. Evaluating these methods require therefore a different approach, in order to identify and emphasize synthesis artifact prone areas, while diminishing the impact, of other types of artifacts, such as those produced by quantization during the video coding. In this paper, we propose an artifact evaluation technique which was shown to provide a better differentiation of synthesis methods, while additional information can be extrapolated from the results about the spatial localization of typical synthesis artifacts and their impact on overall image quality.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.