Flyer and Info Material

• Flyer Virtual Clothing, We developed a dynamic texture overlay method for the visualization of garments. The system allows a user in a virtual mirror or virtual fitting room environment to view himself wearing differently and personalized textured clothes without the need to actually try them on. A single camera captures a moving person and the elastic deformations and illumination of his clothes are estimated. An arbitrary texture is augmented onto the moving garment such that the person seems to wear the virtually textured piece of cloth. In these virtual fitting room environments no additional equipment is required so that the person can move freely in front of the virtual mirror..
• Flyer Virtual Mirror, The Virtual Mirror enhances the visualization of customized consumer articles. Instead of viewing yourself in a real mirror, highly sophisticated 3D image processing techniques are used to verify the appearance of new models. A camera captures the real world and outputs the mirrored images onto a large display which replaces a real mirror. The 3D motion is tracked in real-time and computer graphics models of the consumer articles are augmented into the video such that the person seems to wear the virtual objects.
• Flyer Graphics Streaming, Modern computer games pose high demands on CPU power and are usually only available on high end PCs and game consoles. In order to enable ubiquitous gaming in a home environment, hotel, or café, a 3D graphics streaming solution has been developed. From a server executing the games, the graphics commands are intercepted, encoded and streamed to low cost end devices, where they are rendered locally. The object-based approach offers high visual quality at low delay, crucial for gaming and interactive graphics applications.
• Flyer Text2Video, Text2Video creates video animations with a user selected character that reads a personalized text message. The characters, ranging from synthetic avatars to real people, are generated from a single image. These 3D representations are animated synchronously to the speech synthesized from the text by different text-to-speech engines. Additional motion of the head and the eyes, support for emoticons, and voice modification enhance the visual acceptance of the video. Applications of this technique are SMS to MMS conversion, intelligent user interfaces, or animation services on web sites as well as TV productions.
• Flyer Facial Animation, The Computer Vision and Graphics Group at the Fraunhofer Institute for Telecommunications Heinrich-Hertz-Institute has significant expertise and know how in different aspects of facial animation. We work on facial animation for PCs and mobile devices, avatar control, facial expression analysis, animation from text, speech, or video, creation of 3D head models, and much more. Many different algorithms have been developed and industrial, national and European projects have been conducted. We offer individual solutions in the area of facial animation and to adapt existing components to customer's wishes.
• Flyer MPEG-4 Panoramic Imaging, In contrast to common video, 3D computer graphics environments provide much more interactivity and allow user feedback. Panoramic views, e.g., offer the user the possibility to navigate through virtual scenes. Viewing directions can be changed or details enlarged by zooming into interesting areas. This can be exploited in product presentations for e-commerce, where the user can individually explore the objects from all directions. Also tourism or event management can benefit from the interactivity of virtual guides or walk throughs.
• Flyer 3D Visualization, Many innovative applications like 3DTV, 3D Video, 3d reconstruction, and Free Viewpoint Video require the synchronized and calibrated capturing of multiview video sequences. At the Fraunhofer Institute for Telecommunications, we have therefore worked on different aspects of multiview processing, starting from capturing and calibration to encoding, analysis, and visualization. Different hardware configurations have been investigated and many different algorithms for the analysis of multiview data have been investigated.