Philipp Krähenbühl won the Outstanding Student Paper award at NIPS 2011 for our paper Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials. This year NIPS received 1,400 submissions, 306 of which were accepted for publication. Three of these were picked for the award. (See the conference book for more details.)
I’m an Assistant Professor of Computer Science at Stanford University. I work in visual computing. Current projects in my lab are focused on computer graphics, computer vision, and machine learning.
One stream of research concerns the creation of three-dimensional content. We are developing intelligent 3D modeling tools that can reason about the structure and semantics of complex shapes. These tools support the creation of detailed three-dimensional content without the need for low-level sculpting. I am generally interested in data-driven shape representations and their applications.
Another research direction deals with simulation of human motion and behavior. We are working on high-fidelity physical simulation of human motion based on biomechanical models of the musculoskeletal system. We are also developing new approaches to specifying the behavior of animated characters through demonstrations. I am generally interested in real-time character control.
Other recent and ongoing projects deal with dense probabilistic models for computer vision, with structure-adapting shape deformation, and with animation of gestural communication. In the past I was a theoretician, working in computational geometry. Here is a CV. Old publications are available through DBLP.
Images (click to view)
Videos (click to view)
