Performance comparison of additional versus basic view representation in MPEG Immersive Video

Augmented reality (AR) and virtual reality (VR) enable exciting experiences such as virtual tourism or participation in a live concert from home. These applications are expected to benefit the society as well as being a source of significant revenues. However, such applications are very bit rate- and computation-intensive. The recent MPEG Immersive Video (MIV) format, part of the MPEG-I standard, enables the efficient compression of visual content obtained from several camera feeds by removing redundant information from the various views. This research project aims at investigating the parameters influencing the performance of the MIV format for some uses cases of interest to the industrial partner. This knowledge will enable the partner to select the best MIV format parameters and improve the performance of its products.

Mahshad MahdaviMoghadam
Superviseur universitaire: 
Stéphane Coulombe;Carlos Vazquez
Partner University: