Project Proposals 2011-2012
1)
SEGMENTATION OF TEETH FROM DICOM IMAGE DATA GENERATED BY COMPUTERIZED TOMOGRAPHY
Short Description:
This project falls in the scope of computer-assisted orthodontia. In the long term, we aim
at finding the perfect orthodontic conditions for transforming patient’s teeth arches in a
customized manner. To reach such a goal, orthodontia needs accurate data on teeth
dimensions, shape, position and orientation relatively to reference system of coordinates.
This information is essential to determine the perfect teeth arches shape that will allow
us to develop special techniques to move teeth to the perfect position.
For the ultimate purpose of such a long-term project, we need of a 3D image pipeline to
reconstruct teeth and jaws. This pipeline is fed up by DICOM images as data input, and
includes segmentation and 3D reconstruction of anatomical structures, in particular,
teeth and jaws.
More specifically, this final-year undergraduate project aims to just study and implement
segmentation algorithms for teeth and jaws.
Re-requisites:
- Data structures and algorithms.
- Object-oriented programming.
- Multimedia technologies.
- Computer graphics.
Bibliographical references:
- Cui Jianwen, Fu Bin, and Zhang Han. Study on Three-Dimensional Finite Element
Model of Maxillary Dentition. In Proceedings of the 2011 IEEE/IICME International
Conference on Complex Medical Engineering, May 22 - 25, Harbin, China, 2011.
- Premysl Krsek, Michal Spanel, Petr Krupa, Ivo Marek, and Pavlina Cernochova.
Teeth and Jaw 3D Reconstrucion in Stomatology. In Proceedings of the
International Conference on Medical Information Visualisation - BioMedical
Visualisation (MediViz'2007), IEEE Computer Society, pp.23-28, 2007.
Level: 4 [Easy (1), Average (2), Difficult (3), Very Difficult (4)]
Number of Students: 1
Status: Vacancy filled in by Tiago Ferreira (no. 21549).
2)
3D RECONSTRUCTION OF TEETH FROM DICOM IMAGE DATA GENERATED BY COMPUTERIZED TOMOGRAPHY
Short Description:
This project falls in the scope of computer-assisted orthodontia. In the long term, we aim
at finding the perfect orthodontic conditions for transforming patient’s teeth arches in a
customized manner. To reach such a goal, orthodontia needs accurate data on teeth
dimensions, shape, position and orientation relatively to reference system of coordinates.
This information is essential to determine the perfect teeth arches shape that will allow
us to develop special techniques to move teeth to the perfect position.
For the ultimate purpose of such a long-term project, we need of a 3D image pipeline to
reconstruct teeth and jaws. This pipeline is fed up by DICOM images as data input, and
includes segmentation and 3D reconstruction of anatomical structures, in particular,
teeth and jaws.
More specifically, this final-year undergraduate project aims to just study and implement
3D reconstruction algorithms for teeth and jaws.
Re-requisites:
- Data structures and algorithms.
- Object-oriented programming.
- Multimedia technologies.
- Computer graphics.
Bibliographical references:
- Cui Jianwen, Fu Bin, and Zhang Han. Study on Three-Dimensional Finite Element
Model of Maxillary Dentition. In Proceedings of the 2011 IEEE/IICME International
Conference on Complex Medical Engineering, May 22 - 25, Harbin, China, 2011.
- Premysl Krsek, Michal Spanel, Petr Krupa, Ivo Marek, and Pavlina Cernochova.
Teeth and Jaw 3D Reconstrucion in Stomatology. In Proceedings of the
International Conference on Medical Information Visualisation - BioMedical
Visualisation (MediViz'2007), IEEE Computer Society, pp.23-28, 2007.
Level: 4 [Easy (1), Average (2), Difficult (3), Very Difficult (4)]
Number of Students: 1
Status: Vacancy filled in by Ricardo Ponciano (no. 21421).
3)
ANIMATING FACIAL EXPRESSIONS FOR GAMES AND MOVIES
Short Description:
A facial expression consists of one or more motions or positions of muscles under the skin.
Such muscle movements convey the emotional state of the individual to those with whom
he/she interacts. In this sense, facial expressions work as a form of nonverbal
communication, since they convey social information among humans, though their significance
in the perceiver can vary between cultures.
This work plan must be viewed as part of a wider plan of a project that foresees the
controlled animation of facial expressions. In a first stage, we will use image
acquisition techniques and tools to capture geometric information from human faces.
In a second stage, we will use algorithms of deformable geometry to animate facial
expressions. However, taking into consideration that the stipulated period of time of this
final-year undergraduate project, the project plan will start from a database of geometric
facial expressions encoded as triangular meshes, thus eliminating the first stage of the project.
Re-requisites:
- Data structures and algorithms.
- Object-oriented programming.
- Multimedia technologies.
- Computer graphics.
Bibliographical references:
- Zhigang Deng and Junyong Noh.
Computer Facial Animation: A Survey.
- Junyong Noh and Ulrich Neumann.
A Survey of Facial Modeling and Animation Techniques.
- Shih-Kai Chung.
Facial Animation: A Survey.
- Choong Seng Chan and Flora S. Tsai. Computer Animation of Facial Emotions. In
Proceedings of the International Conference on Cyberworlds (CW'2010),
October 20-22, Singapore, IEEE Computer Society, pp.425-429, 2010 .
- A. Garcia-Rojas, F. Vexo, D. Thalmann, A. Raouzaiou, K. Karpouzis, S. Kollias,
L. Moccozet, and N. Magnenat-Thalmann. Emotional face expression profiles
supported by virtual human ontology. Computer Animation and Virtual Worlds,
vol. 17, pp.259-269, 2006.
- Fabian Di Fiore, Peter Quax, Cedric Vanaken, Wim Lamotte, and Frank Van Reeth.
Conveying Emotions through Facially Animated Avatars in Networked Virtual
Environments. In Arjan Egges, Arnod Kamphuis, and Mark Overmars (eds.),
Motion in Games, Springer-Verlag, pp.222-233, 2008.
Level: 4 [Easy (1), Average (2), Difficult (3), Very Difficult (4)]
Number of Students: 1
Status: Vacancy filled in by Filipe Oliveira (no. -----).
4)
3D RECONSTRUCTION OF POLYHEDRAL OBJECTS FROM 2D ORTHOGRAPHIC VIEWS
Short Description:
3D CAD (computer-aided design) models are usually used by designers and engineers because
of their multiple uses (visualization, simulation, machining). However, multi-view
engineering drawings are still widely used. Accordingly, a conversion tool for obtaining
3D CAD models from 2D drawings is very useful in a broad range of applications.
The interest for this reconstruction problem is witnessed by the large number of research
works published and disseminated in the last three decades. The main object of the present
project is to find out an automatic procedure capable of solving this reconstruction
problem without ambiguities. This algorithm will be limited to polyhedral objects.
Bibliographical references:
- Yary Volpe, Matteo Palai, Lapo Governi, and Rocco Furferi.
From 2D Orthographic Views to 3D Pseudo-Wireframe: An Automatic Procedure.
International Journal of Computer Applications, vol.5, no.6, pp.18–24, August, 2010.
- Tong Lu, Chiew-Lan Tai, Li Bao, Feng Su, and Shijie Cai.
3D Reconstruction of Detailed Buildings from Architectural Drawings.
Computer-Aided Design & Applications, vol.2, no.1-4, pp.527-536, 2005.
Re-requisites:
- Data structures and algorithms.
- Object-oriented programming.
- Multimedia technologies.
- Computer graphics.
Level: 4 [Easy (1), Average (2), Difficult (3), Very Difficult (4)]
Number of Students: 1
Status: Vacancy filled in by ----- (no. -----).