Image Processing CeBIT

The Image Processing department of Fraunhofer HHI presents at CeBIT 2009:

 

Jürgen Rurainsky

Tel. +49 30 31002 267
mobile +49 173 5449185

cebit2009_booth



fhg_booth



Hall 9, booth B 36 (joint booth of the Fraunhofer Society)

(Kopie 15)



(Kopie 17)



(Kopie 18)



3D Broadcasting  - the next Major Step in Home Entertainment

The 3D4YOU project will develop the key elements of a practical 3D television system, particularly, the definition of a 3D delivery format and guidelines for a 3D content creation process. The project generate 3D capture techniques, convert captured content for broadcasting and develop 3D coding for delivery via broadcast, i.e. suitable to transmit and make public. Its objective is to deliver an end-to-end system for 3D high quality media.                

(Kopie 19)



(Kopie 20)



Virtual Mirror - A New Generation of Fitting Rooms

Our "Virtual Mirror" enables a user to view himself wearing virtual clothes without the need to actually try them on. A camera captures a moving person and estimates the clothing’s deformations and illumination. A new virtual texture is augmented onto the garment with correct deformation and shading such that the user seems to wear the virtually textured clothing in the mirror image. The framework combines real video and virtual computer generated content with highly realistic appearance.

cebit2009_booth (Kopie 1)



fhg_booth (Kopie 1)



Hall 9, booth A 30 (THESEUS: joint booth of the Fraunhofer Society)

(Kopie 1)



(Kopie 8)



(Kopie 9)



RUSHES: Multimedia-Search of Audio-Visual Material

The European FP6 research project RUSHES designs, implements and validates a system for indexing, accessing and delivering raw, unedited audio-visual footage. RUSHES promotes the reuse of such material and especially its content in the production of new multimedia assets by offering semantic search capabilities. The core issues are knowledge extraction for semantic-based content annotation, scalable multimedia cataloguing, interactive navigation, non-linear querying and retrieval techniques.

(Kopie 10)



(Kopie 11)



Digitisation and Restoration Safes Cultural Heritage

Within the THESEUS program, text and video restoration algorithms are being developed to achieve optimised digitisation results. The major challenge thereby consists in the automatic identification and correction of erroneous data. For instance, yellowed book pages are optimised for optical character recognition or dirt and scratches are digitally deleted from historical films.

(Kopie 12)



(Kopie 13)



Quality Analysis of Images and Videos

Within the THESEUS program, technologies for quality analysis of images and videos are being developed.  These technologies are capable to detect defects such as dust particles or scratches in images and videos. Furthermore, quality features such as sharpness and contrast of images or videos can be assessed. These hints are used for restoration or for retrieval of high-quality contents in multimedia databases.