Introduction
I am running a research group at the University of Bonn, and I am Adjunct Professor in the Artificial Intelligence Lab at the University of Toronto and TTI-C.
Brief Bio: During 1999-2002 I've been with the MOVI (now PERCEPTION and LEAR) research group at INRIA, where I obtained a doctorate in Computer Science and Applied Mathematics, with a specialization in Imaging, Vision and Robotics. During this period, I collaborated in the EU project VIBES on Video Browsing, Exploration and Structuring. I spent the next two years, 2003-2004, as a research fellow in the AI Lab at the University of Toronto, working on computer vision and probabilistic modeling.
Research: I study problems at the incidence of computer vision, machine learning and computer graphics. I have scientific interests in fundamental non-convex optimization and sampling algorithms and probabilistic modeling, but also in pattern recognition methods for visual scene understanding, in particular: 1) 3d human motion analysis and time series models; 2) object recognition and localization, and 3) shape modeling. Over time, my work has been funded by the EU, The French Ministry of Foreign Affairs, NSERC, NSF and TTI-C.
I am probably better known for the work in 3d human motion reconstruction from monocular video, on which I gave tutorials at IEEE CVPR 2006, ICCV 2005 and the Chicago Machine Learning Summer School, in 2005. I was involved in co-organizing a workshop on Learning, Representation and Context for Human Sensing in Video at CVPR 2006.
Recent Activities and Highlights (see publications for full list)
Greedy Block Coordinate Descent for Large Scale Gaussian Process Regression (L. Bo, C. Sminchisescu), UAI'08
Fast Algorithms for Large Scale Conditional 3D Prediction (L.Bo, C. Sminchisescu, A. Kanaujia, D. Metaxas), CVPR'08
I served as Area Chair for IEEE ICCV'07
People Tracking with the Laplacian Eigenmaps Latent Variable Model (Z. Lu, M. C. Perpinan, C. Sminchisescu), NIPS'07
Support Kernel Machines for Object Recognition (A. Kumar, C. Sminchisescu), ICCV'07
Spectral Latent Variable Models for Perceptual Inference (A. Kanaujia, C. Sminchisescu, D. Metaxas), ICCV'07
Semi-supervised Hierarchical Models for 3D Human Pose Reconstruction (A. Kanaujia, C. Sminchisescu, D. Metaxas), CVPR'07
Generalized Darting Monte-Carlo (C. Sminchisescu, M. Welling), AISTATS'07 (longer TR)
BM3E: Discriminative Density Propagation for Visual Tracking (C. Sminchisescu, A. Kanaujia, D. Metaxas), IEEE PAMI'07
Conditional Models for Contextual Human Motion Recognition (C. Sminchisescu, A. Kanujia, D. Metaxas), CVIU'06
Tutorials on Human Tracking, slides from presentations at CVPR'06, ICCV'05
Learning Joint Top-Down and Bottom-Up Processes for 3D Visual Inference (C. Sminchisescu, A. Kanaujia, D. Metaxas), CVPR'06
Training Deformable Models for Localization (D. Ramanan, C. Sminchisescu), CVPR '06
Conditional Visual Tracking in Kernel Space (C. Sminchisescu, A. Kanaujia, Z. Li, D. Metaxas), NIPS '05
Generative Modeling for Continuous Non-Linearly Embedded Visual Inference (C.Sminchisescu, A.Jepson), ICML'04
Cristian Sminchisescu ▪ Updated October 2007 ▪ Email crismin [at] ces [dot] torontos [dot] edu (replace torontos with toronto)