Perceptually-Optimized 3D Video Streaming Over Next Generation Wireless Home Networks, Код 5 / ИТАЛИЯ



FP7 ICT Call 4 (FP7-ICT-2009-4), Objective ICT-2009.1.5: Networked Media and 3D Internet


Perceptually-Optimized 3D Video Streaming Over Next Generation Wireless Home Networks


3D media is gaining momentum and in the next few years we will probably be surrounded by the most diverse applications and devices pervading many aspects of our lives, ranging from commercial, work, and especially entertainment. However, only recently technology could deliver the tools to produce better immersive experiences by means of auto-stereoscopic and holographic displays at reasonable costs which, with mass production, can be easily expected to become cheaper and more affordable, without needing the user to wear any particular device. On the other hand, wifi home networking is already a widespread reality and the great abundance of broadcasting services allows to foresee more and more 3D media content streaming over the internet.

The ambitious goal of this project is to design the algorithms and the techniques needed to create the infrastructure to provide efficient 3D communication in a wireless home distribution system. On one hand, new metrics to objectively evaluate quality of the perceived audio/visual streams need to be investigated; in fact video quality assessment has always been important and is still a debated issue even for traditional 2D video coding, is essential for 3D video as well to objectively assess the performance and the behavior of coding and transmission techniques.

On the other hand, transmission optimization techniques aimed at minimizing the perceived distortion and improve the performance in terms of the above metrics, need to be developed and tested. This is especially essential when transmission is performed on wireless networks where it is harder to assure a constant level of quality of service, due to losses and delays. Finally, coherently with the home scenario, a prototype using embedded devices will be created.



Currently, the consortium includes six academic as well as industrial partners, all with exclusively technical expertise. However, from the technical standpoint one of the big issues faced by the project is that there are no well-established measures for user's perceived quality in the context of 3D videocommunication, which are indeed fundamental to optimize 3D systems. The concept of “quality” itself in a 3D system is also difficult to define, since it includes picture features such as sharpness etc., as well as more perceptual aspects such as depth perception, fatigue in watching such type of content, etc.

We would like to investigate this important issue together with a partner with a good research background in psychological/perceptual aspects, so that the user's perceptual mechanisms are adequately considered. Together, we expect to:

- define and characterize the main components that affect quality perception in a 3D system;

- on the basis of the previous analysis, propose new quality evaluation metrics that can be implemented in a computer system as well as tested and adjusted by means of subjective experiments (that is, human subjects evaluating video quality in a controlled environment).

The developed measures will then be used as a performance benchmark for the rest of the project. In case the partner decides to join the consortium, the bureaucracy associated with the participation is quite limited: 1-page description of the institution including key persons involved, a short description (1-page) of the activities to be carried out and administrative information about the institution.