Video object coding and rendering for panoramic video

Panoramic video is a simple representation of a dynamic scene or static environment. It allows the users to change their viewpoints interactively in time or along a given path in space. For better users´ experience, it is desirable to incorporate synthetic or real objects into a panoramic video. This paper studies the coding and rendering of these objects in panoramic video using the concept of video objects developed in the MPEG-4 standard. Experimental results showed that better user experience are achieved by incorporating these time varying objects into a static walkthrough application, only at the expenses of slightly more bandwidth and decoding complexity.