Skip to Main content Skip to Navigation

Perceptual quality evaluation of immersive multimedia content : HDR, Light Field and Volumetric Video

Abstract : Immersive multimedia formats emerged as a powerful canvas in numerous disciplines for delivering hyper-realistic user experience. They can take many forms, such as HDR images, Light Fields, Point Clouds, and Volumetric Videos. The goal of this thesis is to propose novel methodologies for the quality assessment of such multimedia content. The first part of the thesis focuses on subjective image quality assessment. More specifically, we propose a content selection strategy, observer screening tools, and an extensive analysis on the reliability of crowdsourcing platforms to produce a large-scale dataset. Our findings improve the reliability of the collected subjective annotations and address issues to transfer laboratory experiments into crowdsourcing. The second part contributes to the objective quality evaluation with a learning-based image quality metric utilizing the just noticeable difference information and a no-reference light field image quality metric based on epipolar plane image representations. Finally, we investigate the impact of temporal pooling methodologies in objective quality metric performances for volumetric videos. Overall, we demonstrate how our findings can be used to improve the optimization of processing tools for immersive multimedia content.
Complete list of metadata
Contributor : Ali Ak Connect in order to contact the contributor
Submitted on : Thursday, May 19, 2022 - 9:17:56 AM
Last modification on : Saturday, June 25, 2022 - 3:04:29 AM


Files produced by the author(s)


  • HAL Id : tel-03672037, version 1


Ali Ak. Perceptual quality evaluation of immersive multimedia content : HDR, Light Field and Volumetric Video. Computer Science [cs]. Nantes Université, 2022. English. ⟨tel-03672037⟩



Record views


Files downloads