| Address: | Institut für Numerische Simulation Wegelerstr. 4 (Flachbau) 53115 Bonn Germany |
| Office: | We4 0.022 |
| Phone: | +49 228 733522 |
| E-Mail: | cristian.sminchisescu ins.uni-bonn.de |
The field of artificial intelligence emerged at the intersection of mathematics, physics, computer science and cognitive science, with the quest of creating computer programs that learn from experience. The challenge is to build robots able to perceive, see the real world, and act intelligently in unexpected, uncertain circumstances, as humans do. Computational visual perception and motion capture technologies started more than a century ago with Helmholtz's mathematics of the eye and the emerging photographic techniques. Nowadays recognizing visual objects, understanding video content and transferring this into 3-dimensional models is the basis of advanced special effects, digital libraries and image indexing systems, or the construction of humanoid robots that can localize objects, recognize people, comprehend actions and interact with the world seamlessly. Our research group at the Faculty of Mathematics and Natural Science at the University of Bonn develops new computational methods for artificial intelligence and visual recognition, aiming at creating programs able to understand the visual world from images and at building mathematical models that can be fitted into the `mind' of an intelligent agent.
Our research group's team (Carreira, Ion, Li, Sminchisescu) is a winner of the PASCAL VOC 2011 Object Class Segmentation Challenge. Talk available here.
C. Sminchisescu will lecture at the ETHZ11 Computer Vision Summer School, Zurich.
October 2010: Paper on Random Fourier approximations for skewed multiplicative histogram kernels by Li, Ionescu and Sminchisescu received DAGM prize. Code available here.
September 2010: Our group’s team (Carreira, Li, Sminchisescu) is the joint winner in the PASCAL VOC 2010 Object Class Segmentation Challenge. See our segmentation and recognition methods. The new segmentation method pursued by our second-scoring team (Carreira, Ion, Sminchisescu) is available here. Our CPMC segmentation code is available here. Our most recent result is 43.8%, see the table for individual class scores.
March 2010: We have obtained state-of-the art results in the HumanEva benchmark, for monocular 3D human pose estimation. See our structured prediction and mixture of experts methods. Code available. See also the IJCV 2010 Special Issue: Evaluation of Articulated Human Motion and Pose Estimation.
C. Sminchisescu will give a tutorial on Structured Prediction at 3DPTV 2010, Paris. Slides are available here.
C. Sminchisescu will lecture at the ETHZ10 Computer Vision Summer School, Zurich.
September 2009: Our group’s team (Carreira, Li, Sminchisescu) scored the winning entry in the PASCAL VOC 2009 Object Class Segmentation Challenge. See talk here.
July 7-11th, 2008, C. Sminchisescu lectures at the AERFAI Summer School on New Trends in Pattern Recognition for Motion Analysis (PRMA'08), Barcelona.
March 2008: Group's coverage in the University Press.
February 2007: The group becomes part of the Hausdorff Center of Excellence.
Our group's research is supported, in part, by a Marie Curie Excellence Grant to C. Sminchisescu.
June 2006: Workshop on Learning, Representation and Context for Human Sensing in Video at IEEE CVPR, New York.
June 2006: Tutorial on Tracking People at IEEE CVPR, New York.
