PhD position at LIRIS, University of Lyon :Visual quality metrics for complex 3D scenes in immersive environments

Location : LIRIS, Lyon, France – http://liris.cnrs.fr/

Advisors : Guillaume Lavoué – http://liris.cnrs.fr/guillaume.lavoue/ –LIRIS Lab, University of  Lyon

. Patrick Le Callet -www.irccyn.ec-nantes.fr/~lecallet–LS2N lab, University of Nantes.

Context:

Three-dimensional (3D) graphics are commonplace in many applications such as digital entertainment, cultural heritage, architecture and scientific simulation. These data are increasingly rich and detailed; as a complex 3D scene may contain millions of geometric primitives, enriched with various appearance attributes such as texture maps sesigned to produce a realistic material appearance, as well as animation data.

The way of consuming and visualizing this 3D content is now evolving from standard screens to Virtual and Mixed Reality (VR/MR). However, the visualization and interaction with 6 degrees of freedom with large  and  complex  3D  scene  remains  an  unresolved  issue  in  such  immersive  environments,  especially when the scene is stored on a remote server. Two distinct bottlenecks exist: (1) the potential complexity of  a  3D  scene  that  can  be  displayed  to  the  user  on  a  VR/MR  head-mounted  display  is  substantially smaller than for a standard screen, because the GPU must generate 4 times more images (to ensure two images  per  frame  and  a  sufficient  frame-rate  to  prevent  motion  sickness);  (2)  since  an  increasing number of VR/MR applications consider 3D data stored on remote servers, strong latency problems may be encountered, caused by the streaming of the scene on the display device.

The  proposed  PhD  position  is  funded  by the ANR  PISCo  project  (Perceptual  Levels  of  Detail  for Interactive  and  Immersive  Remote  Visualization  of  Complex  3D  Scenes)  which  aims  at  proposing novel algorithms and tools allowing inter active visualization, in these constrained contexts (Virtual and Mixed reality,  with  local/remote  3D  content),  with  a  very  high  quality  of  user  experience.  As  3D  scenes  are visualized through a certain viewport, we seek to optimize the display in this view port by proposing (1) Tools  for  the  generation  and  compression  of  high  quality  levels  of  details,  (2)  Visual  quality  metrics capable  of  predicting  the  quality  of  these  levels  of  detail  and  driving  their  generation,  (3)  Visual attention models capable of predicting where the observer is looking and thus selecting and filtering the primitives and levels of detail.

A distinctive property of the project lies into the fact that we will consider rich 3D data, including not only geometric information but also animation and complex physically based materials represented by texture maps (color, metalness, roughness, normals).

The ANR PISCo project is funded by the French Research Agency and involves three academic partners: LIRIS (University of Lyon), LS2N(University of Nantes) and INRIA TITANE (Sophia-Antipolis).

Subject:

The proposed  PhD position  concerns  the  item  (2)  above.  The  objective  is to study  the perceptual mechanisms involved in immersive 3D data visualization, where 3D data do not contain only geometry  but  also  complex  material  (different  kinds  of  texture  maps),  illuminationinformation  and animation  data.  The  final  objective  is  to  produce  both  near-threshold  and  supra-threshold quality indices,  capable  of  predicting  the  perceptual  impact  of  modifications applied  to  the  geometry, texture maps and animation of a rich 3D scene.

The PhD student will benefit from the recent related works from the involved research teams:

L Krasula, P Le Callet, K Fliegel, M Klíma, Quality Assessment of Sharpened Images: Challenges, Methodology, and

Objective Metrics, IEEE Transactions on Image Processing 26 (3), 1496-1508, 2017.

Guo,  Vidal,  Cheng,  Basu,  Baskurt,  Lavoué,  Subjective  and  Objective  Visual  Quality  Assessment  of  Textured  3D Meshes, ACM Transactions on Applied Perception, Vol. 14, No. 2, Article 11, October 2016.

  1. Lavoué, G.,  M.C.  Larabi,  L.  Vása,  On  the  Efficiency  of  Image  Metrics  for  Evaluating  the  Visual  Quality  of  3D Models, IEEE Transactions on Visualization and Computer Graphics, vol. 22, n°8, p. 12, 2016.
  2. Nader, K.  Wang,  F.  Hétroy-Wheeler,  and  F.  Dupont,  Just  noticeable  distortion  profile  for  flat-shaded  3D  mesh surfaces. IEEE Transactions on Visualization and Computer Graphics, vol. 22, n°11, pp. 2423-2436, 2016.
  3. Narwaria, M.  Perreira  Da  Silva,  P.  Le  Callet. HDR-VQM:  An  objective  quality  measure  for  high  dynamic  range video. Signal Processing: Image Communication, 2015, vol. 35, p. 46-60.

 Required qualifications:

Master’s or Diploma degree in Computer Science, strong experience  with  C++ programming, good knowledgeof image processing and computer graphics.

Duration:

3 years.

Monthly net salary:

≈ 1550€.

Possibility to teach starting the 2ndyear of PhD(+200€ / month).

Starting date:

No specific constraint, the PhD can start from November 2017 to June 2018.

Contacts:

glavoue@liris.cnrs.fr

patrick.lecallet@univ-nantes.fr

Catégorie(s) : Emploi et carrière, Offres de thèses

Laisser un commentaire

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *

*


9 × = soixante trois