TU Delft
print this page print this page     
2016/2017 Architecture Master Geomatics
3D Modelling of the Built Environment
Responsible Instructor
Name E-mail
Dr. S. Zlatanova    S.Zlatanova@tudelft.nl
Name E-mail
Dr.ir. P. Nourian    P.Nourian@tudelft.nl
Contact Hours / Week x/x/x/x
18 lectures
24 labs
98 selfstudy
Education Period
Start Education
Exam Period
Course Language
Expected prior knowledge
Knowledge of programming in at least one language (e.g. MATLAB, Java, C#, or Python) is required. The course GEO1000--Python Programming should have been passed prior to this course. Preliminary knowledge of Linear Algebra is recommendable.
Course Contents
In this course, students will learn the fundamentals of 3D modelling of the built environment. This course covers the process of deciding what kind of model needs to be made for representing different phenomena in 3D, selecting an appropriate data modelling approach, applying suitable data processing algorithms, and data management for the purpose of extracting information by spatial analysis, modifications, conversions, and explorations needed to create 3D models for different built environment modelling applications.

This course focuses on 3D geometry, topology and semantics of 3D models. All subjects are covered both theoretically in lectures and practically through a series of hands-on labs (involving algorithm design and programming) focused on assignments (e.g. on 3D reconstruction of building models from point clouds).

The topics covered by this course can be subdivided into four large groups as below:
1. 3D representations such as B-reps, Voxels, and parametric freeform curves and surfaces (e.g. NURBS) in BIM, CAD, or GIS 3D models (e.g. CityGML)
2. 3D data models and data structures for modelling Geometry, Topology and Semantics of human-made objects in the built environment (e.g. buildings) in 3D Geo-DBMS
3. 3D Analytic Geometry, Linear Transformations, and Intersections in 3D required for fundamental spatial algorithms
4. 3D reconstruction methods using point cloud data sets: e.g. by estimating normal vectors, curvatures, classification, segmentation, object reconstruction.
Study Goals
After the course the student will able to:
1. analyse advantages and disadvantages of different 3D models (geometry, topology, semantics) and advise on representations and their level of detail;
2. describe, explain, evaluate, and apply processing algorithms on one or multiple data models, and justify the approach;
3. analyse, use, and estimate benefits of different 3D reconstruction approaches with respect to the desired richness of final model in terms of objects, attributes, relationships, semantics;
4. evaluate approaches and tools for storage and exchange of 3D models (data formats and databases) for different applications, decide on an approach and justify the decision; and
5. generalise their knowledge of 3D modelling to solve complex problems in the domain area of built environment
Education Method
Lectures: 20 hours; 9 regular lectures (18 hours) and 1 invited lectures (2 hours)

Practical: 18 hours; 3-4 group assignments (2-3-4 students). The group assignments are on data modelling, database storage and algorithms for 3D modelling using Rhino and Grasshopper as visualization environments for programming.

Self-study: 104 hours
Literature and Study Materials
Selected book chapters, scientific and professional articles, Rhino and Grasshopper software applications, freeware Dot NET or Python libraries for Scientific Computing (Linear Algebra)
Written exam and practice work. The written exam result will determine 40% of the final score (including a mid-term quiz worth 15%), the labs 60%.
Period of Education
Quarter 3
Course evaluation
For the course evaluations see: http://kwaliteitszorg.bk.tudelft.nl/