You are here

Rendering of HDR 3D points cloud

Team and supervisors
Department / Team: 
PhD Director
Remi COZOT
Co-director(s), co-supervisor(s)
Olivier LE MEUR
Contact(s)
PhD subject
Abstract

Context: ANR REVERY Project

In the last two decades, industries and researchers proposed significant advances in media content acquisition systems in three main directions:
•    Increase of resolution and image quality with the new ultra-high-definition (UHD) standard;
•    Stereo capture for 3D content; and
•    High-Dynamic Range (HDR) imaging.
Existing technologies and methods, suffer from the lack of a joint representation of the different media formats discussed above. The REVERY project aims at providing solutions to replace the traditional video media (mono- or stereoscopic) by a richer data stream: HDR 3D points cloud stream – 3D HDR PC. This richer media will jointly embed UHD, HDR and Depth information in a dedicated format. The project will
•    Develop a demonstrative prototype of a dedicated acquisition system, and
•    Prove their benefits to media creation industry.
The REVERY project will enable versatile postproduction and re-usability of single shootings of live action under less-restrained range of lighting intensities (e.g., outdoor scenes). Its outcome is important with a twofold impact: (i) societal (for consumers and experts), by increasing content quality and (ii) economical, by reducing production costs. Ultimately, the creation approach will evolve to provide a completely new experience of screen viewing to the public.

Objectives:

The objective of the PhD is the development of software component, called 3D vTMo (3D video Tone Mapping operator) that renders a 3D HDR PC with meta-data into images that are compliant with the targeted screen features. Indeed, so far none of the existing screens (HDR or stereoscopic or auto-stereoscopic) is able to display videos that are 3D, HDR, and UHD. Many display manufacturers just announced that UHD-HDR screens would come onto the market shortly. However, stereoscopic or auto-stereoscopic screens should remain only LDR compatible in the near future. The main functions of our 3D vTMo are:

(1)    Computation of several views from the 3D HDR point cloud,
(2)    Reduction of the high dynamic range to low dynamic range,
(3)    Reduction of the depth range to avoid visual fatigue, and
(4)    Taking into account meta-data such as the focus plane or the light aesthetics, etc.

Even if many approaches have been proposed for video tone mapping, this latter is still challenging. Indeed many artifacts [1, 2] can occur when performing the tone mapping. Built on our previous experience, a new tone mapping software will consist of two different components:

(1)    Video tone mapping (to display HDR contents on LDR screens) and
(2)    Stereo video tone mapping (to enhance the perceived luminance range on LDR stereoscopic screens).

The new tone mapper will make use of the rich information (3D + HDR) to ensure spatial and temporal coherency while avoiding flickering artifacts. Some other features will be added such as white balance.

Bibliography

[1] R. Boitard, D. Thoreau, K. Bouatouch, and R. Cozot, “Temporal Coherency in Video Tone Mapping, a Survey,” in HDRi2013 - First International Conference and SME Workshop on HDR imaging (2013),, 2013, pp. 1–6.

[2] R. Boitard, D. Thoreau, R. Cozot, and K. Bouatouch, “Survey of temporal artifacts in video tone mapping,” in HDRi2014 - Second International Conference and SME Workshop on HDR imaging (2014), 2014, pp. 1–6.Page 1 of 1

Place: 
IRISA - Campus universitaire de Beaulieu, Rennes