Computer scientists work to make interactive 3D graphics part of the Web
07 August 2012
Interactive 3D graphics are not available on the Internet even though almost all PCs, as well as mobile and embedded devices, have the 3D graphics hardware to process it.
Car Design in 3D: Philipp Slusallek simplifies interactive 3D graphics on the World Wide Web (photo credit: bellhäuser - das bilderwerk)
Now, computer scientists from the German Research Center for Artificial Intelligence and Fraunhofer Institute for Computer Graphics Research are joining forces to change that. Together, they are working to describe computer scenes in spatial detail directly within the websites’ code.
The German Research Center for Artificial Intelligence (DFKI) in Saarbrücken and the Fraunhofer Institute for Computer Graphics Research (Fraunhofer IGD) in Darmstadt have agreed on a common proposal to extend the Hypertext Markup Language (HTML) to also include advanced graphics capabilities.
The proposal allows for easily describing 3D geometry, its material properties (shaders), lights, and virtual cameras as new HTML elements. "These elements can be used almost exactly like existing HTML elements. Therefore, millions of Web developers can quickly begin to include interactive 3D visualisations in their Web applications," says Philipp Slusallek, professor for computer graphics at Saarland University and scientific director in the DFKI and in the Intel Visual Computing Institute (VCI).
"By identifying a small core of essential building blocks from the existing prototypes and scene-graph standards for a interactive three-dimensional experience on the Web, we have made it as simple as possible for browser vendors to include the new technology but still offer Web developers the full flexibility for designing fully dynamic and interactive 3D Web experiences," says Johannes Behr, head of the Competence Center for Visual Computing System Technologies at Fraunhofer IGD and leader of the X3DOM development team.
The proposed declarative extension to HTML offers a high-level approach for Web developers, extending the low-level procedural approach to talk to graphics hardware now offered by WebGL. "Instead of requiring Web developers to become WebGL experts or learn new APIs, we are adding advanced graphics capabilities to HTML, allowing Web developers to reuse their existing skills and directly apply the Web technology they use on a daily basis," explains Kristian Sons, head of the XML3D research group at DFKI/VCI.
At least two implementations will be made available: For the short term a JavaScript implementation using WebGL for rendering will allow developers to get started immediately, while a native implementation integrated within the browser will offer optimal performance and full functionality.
The two institutes originally each had their own proposals, namely X3DOM and XML3D, but recently joined forces to define a common standard. The two groups distilled their technologies into the essential components needed for bringing interactive and highly dynamic 3D graphics to the declarative world of HTML.
"Our two proposals were essential in order to gain experience and evaluate a number of different approaches," says Yvonne Jung, senior researcher and core developer for X3DOM at Fraunhofer IGD. This joint research was supported by the German Software-Cluster initiative and the Intel Visual Computing Institute at Saarland University.
The joint proposal will be officially presented at the SIGGRAPH 2012 and Web3D conferences this week in Los Angeles.