Skip navigation
Skip navigation
You are using an outdated browser. Please upgrade your browser.

Televirtual: Press Release – April 24rth 2003.

Meet TOMMI: the best a girl can get…
UK Technology company Televirtual has unveiled what its claiming as the world’s most advanced virtual TV broadcaster.
Tommi has been developed specifically to showcase broadcast TV potential within Televirtual’s award-winning RAP real-time production software.

She was demonstrated in London yesterday to an invited audience of TV producers and programme developers.
Created by Televirtual artists and programmers to exploit the latest PC graphics rendering hardware from industry leaders, nVidia, Tommi was constructed in Alias Maya software, and then exported via a bespoke application (which preserves the high-end functionality of Maya) into the RAP real-time system.

Tommi is a big girl. The surface mesh sitting over her animation skeleton is made up of some 70,000 polygons. Only a few weeks ago, such a complex model running at production frame rates in a PC environment would have been impossible. But the latest Geeforce FX graphics chipsets from nVidia not only make her a reality, they also allow complex real-time shading and lighting routines (including 3D surface texturing) which give her a level of natural realism unmatched by any other live broadcast animation.

Tommi is an interactive character so her image must be generated in real time. For TV, this means that she has to be drawn at least 25 times a second. This is possible only by using the latest in graphics hardware and programming techniques, with many shading programs executed on the graphics card. These shaders are used to determine the colour of each pixel used in the final image. Her skin shader, for example, models the effects from three lights in the scene and how these fall on to her skin and are reflected back to the eye. It takes into account the effects of shadowing, the sheen and the texture of a person's skin

Tommi can interact with virtual or real-world scenes in real-time. She can be driven by motion-captured animation files, or key-framed animation scripts, all under simple keyboard control. Interfaced to natural-language speech technologies, she can generate real-time lip-synchronised speech animation, to accompany live, recorded or computer-generated voices.

Facial animation and expressive gestures can be generated or triggered through real-time markerless facial tracking systems such as Image Metrics’ Real Face’. Tommi can play-back pre-designed facial gestures built as a series of morph targets, or, using her complex facial structure of bones and muscles, she can perform intuitive and unique gestures captured from a live actress in real time.

For images, contact Tim Child, 01603 767493,

This press release was distributed by ResponseSource Press Release Wire on behalf of Intermedialab Ltd in the following categories: Consumer Technology, Personal Finance, Business & Finance, Computing & Telecoms, for more information visit