My research focuses on fundamental methods to generate and manipulate images using computers. We develop algorithms and systems for realistic and real-time rendering, and animation and modeling of three-dimensional shapes. We are also interested in novel representations for 3D geometry, such as point-based representations. Finally, we investigate signal processing techniques, in particular for multi-view 3D displays. My research has applications in virtual reality, digital entertainment, multimedia, and data visualization. Visit my list of publications for a detailed overview of our published work, or follow the links below to individual projects.
Rendering, or image synthesis, is a core problem in computer graphics. We develop algorithms for efficient, physically-based rendering. We are also interested in rendering for interactive applications, and we investigate techniques to simulate light transport in real-time. Our research strives to finally make realistic Monte Carlo rendering practical in a broader range of real world application scenarios. We have contributed to a variety of issues with a series of publications in ACM Transactions on Graphics, including filtering noise textures, efficient multidimensional adaptive sampling techniques, novel approaches to photon mapping, rendering participating media, and flexible, meshless precomputed radiance transfer.
Recently, our work on image space denoising and adaptive rendering has had a strong impact in the research community and in industry, because it effectively reduces noise while easily interfacing with conventional rendering systems. We are licensing our work to Innobright Technologies Inc., and other players in the industry have adopted techniques inspired by our research. We led the community by organizing an ACM SIGGRAPH 2015 course on denoising in Monte Carlo rendering.
We are currently leading a research effort on an innovative new Monte Carlo rendering technique called gradient-domain path tracing, which we presented at ACM SIGGRAPH 2015. We have shown that this approach can significantly reduce the required computation time compared to conventional techniques, and believe that it will become a key building block of the next generation of rendering algorithms. Gradient-domain rendering has opened a significant number of research opportunities to further improve the approach, which we are currently pursuing in a project funded by the Swiss National Science Foundation. We are also excited to explore the implications of novel display technologies, such as AR and VR goggles or multiview 3D displays, on signal processing and realistic rendering algorithms. We believe real-time denoising techniques will play a key role to make realistic rendering practical in AR and VR applications, and we continue to investigate and adapt such techniques for these scenarios.
A fundamental challenge for computer graphics applications is the current modeling bottleneck, which means that the process of modeling three-dimensional geometry, surface appearance, animated characters, etc. is time consuming, often manual, and expensive. We are pursuing a research agenda based on data-driven modeling to address this issue, which is the powerful concept of using data acquired from the real world or existing models to drive the modeling process.
We have contributed innovative techniques for data-driven modeling of hair, 3D geometry, textures, and stereo images, all appearing in ACM Transactions on Graphics publications. We have recently been awarded an ERC Consolidator Grant for our project “D2Graphics: Data-driven Modeling for Computer Graphics" to further investigate our vision of data-driven modeling.
In a related project with my colleague Prof. Favaro and sponsored by the Swiss National Science foundation, we are currently investigating techniques for sketch-based image synthesis that also leverage and further develop state-of-the-art deep learning techniques. We are exploiting the availability of internet-scale image databases to enable sketch analysis, sketch-to-image matching, and image-from-sketch synthesis.
3D shape representations form the foundation for most higher level applications, and we are investigating new 3D representations and 3D geometry processing techniques that are suitable for these tasks. In our early research we pioneered point-based techniques for reconstruction, rendering, and editing of 3D geometry, which have inspired a whole new research area, and we continue to contribute state of the art techniques for processing point-sampled 3D data.
We consider the development of 3D shape representations that support intuitive interactive modeling one of the key problems in geometry processing. We have introduced the idea of example-based mesh deformation, which we called mesh-based inverse kinematics. Recently, the proliferation of inexpensive, real-time RGB+depth video cameras has offered a boon of new research opportunities. Our vision is to exploit this data to automatically construct 3D models that mimic the properties of their real-world counterparts, such as the degrees of freedom of articulated objects.
Automultiscopic displays show stereoscopic images that can be viewed from any viewpoint without special glasses. They hold great promise for the future of television and digital entertainment. We develop signal processing techniques to optimize image quality by reducing sampling artifacts and adapting the signal to the display properties. We are also interested in multi-view content creation and manipulation techniques.
I am also intrigued by applications of algorithms and numerical techniques from Computer Graphics to other scientific areas. For example, in a collaboration with Michel Milinkovitch, a biologist at the University of Geneva, we have been investigating the mechanisms behind structural and color patterns on animal skins.
Noise, or variance, is a fundamental problem in Monte Carlo rendering, and we have contributed several algorithms for effective denoising of Monte Carlo renderings, hence significantly reducing the required rendering time. This has inspired us to also consider the general image denoising problem. With Claude Knaus, a former PhD student, we developed the dual-domain filter, a state of the art denoising filter that can be implemented in only a few lines of Matlab code. We are currently also investigating image restoration techniques using deep convolutional neural networks.