In video processing 3D video is the next revolution and can be considered as comparable evolution, in video technology, after the introduction of colored video systems. The principles of 3-D video or stereopsis are abundant in nature and virtually all animal have stereo vision system because of two eyes. Depending on the position each of two eyes of viewer sees slightly different image of the same scene. Hence, the viewers' left eye receives a signal that is different from what his right eye gets; this gives, provided the signals have been properly processed, the impression of depth. New auto-stereoscopic products use an image-plus-depth interface and multi view auto-stereoscopic displays are making their way in the consumer electronic market. In portable embedded consumer systems market (Mobiles, PDA, e.t.c) the overwhelming acceptance of 4G communications protocols provides the necessary communication infrastructure for 3D video conferencing. Necessary infrastructure is there but incorporation of 3D video in portable embedded devices poses gigantic problems such as computational cost, reliable algorithms for disparity extraction and management of quality of service e.t.c.

The entertainment industry is increasingly accepting the 3D and beyond video technologies to provide more viewing flexibility and realism to user. A growing number of 3D productions from the entertainment industry use a stereo format but 3D video view is only possible if device is capable of this.

In this seminar we will focus on the issues related to stereo video processing such as methods to extract depth map, 3D video compression (encoding/decoding) and transmission issues including the error resilience/error recovery e.t.c.

Tutor: M. Shafique (Tel. 608-45732), M. Khan (Tel. 608-47639).

Language: Englisch

Meeting schedule: will be informed by email.

Application: via registration form.