Ioan Cleju

Doctoral Student in the PhD program since 01.01.2005.
May 2, 2006 - July 31, 2006: on leave at McMaster University, Hamilton, Ontario, Canada, with Prof. Dr. Xiaolin Wu. Octobre, 2008 - Graduation

advisors

1. Prof. Dr. Dietmar Saupe
2. Prof. Dr. Ulrik Brandes

organisational data

Room	:	Z 713
Tel.	:	+49 (0)7531 / 88-2200
E-mail	:	firstname.lastname "at" uni-konstanz.de
Other Resources	:	http://www.inf.uni-konstanz.de/~cleju/

project description

Perceptual Distances and Texture Registration for 3D Models

Quantifying the perceived dissimilarity between 3D models and their simplified copies is necessary for assessing the quality of simplification and compression algorithms, and for Level-Of-Detail (LOD) management. This thesis examines several strategies used for the evaluation of objective distances with respect to user studies. We show that ordinal analysis on the LOD sequences does not provide enough data to differentiate between several objective distances. Popular evaluation strategies, such as based on ratings, provide however ordinal data. We propose a new experimental setup that allows parametric evaluation of objective distances with respect to the user study. The case study included six objective distance measures and we found that all image-based distances were better than those geometric-based.

A second topic covered in this thesis is texture registration for 3D models. The common 3D acquisition pipeline considers geometry acquisition (by 3D scanning) and texture acquisition (by photographing) as two independent steps. The texture registration solves the 2D-3D mapping problem by recovering the parameters of the photo cameras. Commonly, the patches of the model are visible in several images. We propose to use this additional information to improve the registration algorithm by adding corresponding objective functions. We define objective functions based on mutual information between each image and the surface model and between each pairs of images that sample a common patch of the surface. The mutual information has several advantages over other registration criteria, including that it is robust and does not need preprocessing and feature extraction. In various experiments we showed that the extended optimization approach is more robust with respect to the initialization and leads to increased accuracy of registration.

publications

The following list of publications covers only those, which are or were published during participation at the Graduiertenkolleg / PhD program.

Articles in Journals

2010

Cleju, I., Saupe, D., Evaluation of texture registration by epipolar geometry, The Visual Computer, Vol. 26, No. 11, pp. 1407-1420, 2010. Link abstract

Conference Papers

2009 2007 2006 2005

2009		Zheng, H., Cleju, I., Saupe, D., Highly-automatic MI based multiple 2D/3D image registration using self-initialized geodesic feature correspondences, Asia Conference on Computer Vision (ACCV), Lecture Notes in Computer Science (LNCS), Zha, H. (ed.) , pp. 426-435, 2009, Springer-Verlag. abstract
2007		Cleju, I., Saupe, D., Stochastic optimization of multiple texture registration using mutual information, Annual Symposium of the German Association for Pattern Recognition (DAGM), Lecture Notes in Computer Science (LNCS), Vol. 4713, pp. 517-526, 2007, Springer-Verlag. abstract
2006		Cleju, I., Saupe, D., Evaluation of supra-threshold perceptual metrics for 3D models, Symposium on Applied Perception in Graphics and Visualization (APGV), pp. 41-44, Boston, USA, July 2006. abstract
2005		Cleju, I., Fränti, P., Wu, X., Clustering by principal curve with tree structure, International Symposium on Signals, Circuits and Systems (ISSCS), Vol. 2, pp. 617-620, July 2005. abstract Cleju, I., Fränti, P., Wu, X., Clustering based on principal curve, Scandinavian Conference on Image Analysis (SCIA), Lecture Notes in Computer Science (LNCS), Vol. 3540, pp. 872-881, June 2005, Springer-Verlag. abstract

Phd Theses

2008

Cleju, I., Texture Registration for 3D Models, University of Konstanz, July 2008. abstract

awards

Award for The Best Pattern Recognition Master's Thesis in Finland in 2004 by The Pattern Recognition Society of Finland.

curriculum vitae

2003 - 2004	Studies of Computer Science at University of Joensuu, Finland Degree: M.Sc.
2001 - 2002	Working at Isratech Romania as asic designer.
1998 - 2003	Studies of Computer Engineering at Technical University Gh. Asachi, Iasi, Romania Degree: Diploma Engineer.

	GK Home	Department	GK Internal	FAQ	Contact			Software
	People	Research	Teaching	Events	Publications	Applications	Gallery	PhD Colloquium