Rendering is an advanced field of computer graphics and offers impressive possibilities, as can be observed in the cinemas. The development that CGI permeates so many movies shows that there is an increasing need for synthetic pictures which cannot be distinguished from real footage. This also means that the audience has a trained eye to detect remaining rendering artifacts, so modern rendering systems need to eliminate as much of those as possible. This talk shows which techniques can be applied to today's rendering challenges, including difficult lighting, complex materials, massive geometry, and sub-millimeter detail which is necessary to create believable CGI characters.
Carsten Dachsbacher is Full Professor at the Karlsruhe Institute of Technology and Head of the Institute for Visualization and Data Analysis. His research focuses on high performance graphics, (interactive) global illumination, scientific visualization and perceptual rendering, on which he published several articles at various conferences and journals including SIGGRAPH, EG, EGSR etc. He has been a tutorial speaker at SIGGRAPH, Eurographics and the Game Developers Conference and reviewer for various conferences and journals.
Johannes got his PhD in media informatics from Ulm university in 2011. After that he worked as a researcher for Weta Digital in Wellington, New Zealand. Since 2013 he is back in Germany and works as a post-doctoral fellow at the Karlsruhe Institute of Technology.
KIT Computer Graphics Group
The Chair of Computer Graphics at the Karlsruhe Institute of Technology conducts research on the analysis, processing, and display of visual information. The goal is to render and manipulate synthetic images efficiently and interactively, to create convincing computer animations, and to provide insightful visualizations of complex data.
These tasks require among others models for describing virtual scenes, representations for creating and manipulating scenes, and methods for image synthesis (rendering). Visualization further requires a preparation of data and mapping it to well-interpretable graphical and visual primitives. Important aspects in all these fields are efficiency, interactivity, and consumption of resources.
Our projects and research include among others:
- (interactive) photorealistic rendering and global illumination
- real-time rendering (video games, simulators, ...)
- general purpose computation on GPUs
- visualization of complex and large data
- perceptual methods
- simulation of light and sound propagation in virtual scenes
- interactive manipulation of images and 3D scenes
- procedural modeling