Research Group of Prof. Dr. C. Sminchisescu
Mathematical Sciences



Dipl.-Inform. Catalin Ionescu

Address: Institut für Numerische Simulation
Wegelerstr. 4 (Flachbau)
53115 Bonn
Germany
Office: We4 0.023
Phone: +49 228 732721
E-Mail: catalin.ionescu.ins.uni-bonn.de

Publications

[1] C. Ionescu, F. Li, and C. Sminchisescu. Latent Structured Models for Human Pose Estimation. In IEEE International Conference on Computer Vision, November 2011.
bib | .pdf 1 ]
[2] C. Sminchisescu, L. Bo, C. Ionescu, and A. Kanaujia. Feature-based Human Pose Estimation, in Guide to Visual Analysis of Humans: Looking at People. Springer, 2011.
bib ]
[3] F. Li, C. Ionescu, and C. Sminchisescu. Random Fourier Approximations for Skewed Multiplicative Histogram Kernels. In Lecture Notes for Computer Science (DAGM), September 2010. DAGM paper prize.
bib | .pdf 1 ]
[4] C. Ionescu, L. Bo, and C. Sminchisescu. Structural SVM for Visual Localization and Continuous State Estimation. In IEEE International Conference on Computer Vision, September 2009.
bib | .pdf 1 ]
[5] C. Ionescu and C. Sminchisescu. Hierarchical Latent Variable Models for Human Pose Inference. Snowbird Learning, April 2009.
bib | .pdf 1 ]
[6] C. Sminchisescu and C. Ionescu. Hierarchical Spectral Latent Variable Models. Snowbird Learning, April 2008.
bib | .pdf 1 ]

Bio

I am a doctoral student in the Computer Vision and Machine Learning group of the Institute for Numerical Simulation (University of Bonn) under the supervision of Prof. Cristian Sminchisescu. As part of DFH-UFA I have received a Diplome d'ingenieur from the Institut National de Sciences Appliquees de Lyon and a german Diplom fur Informatik from the University of Karlsruhe. My master thesis was on program verification under the supervision of prof. Peter Schmitt and was concluded as part of an Interact Scholarship while at School of Computer Science, Carnegie Mellon University in prof. Edmund M. Clarke Jr.'s group.

Interests

At the moment I am looking into latent variable models, dimensionality reduction and related methods, as well as their applications in computer vision problems.