3D Data Processing & Visualization

LEARNING OUTCOMES

Basic purpose of this course is the familiarization of students with the prevailing methods/techniques for collection of 3D spatial information (automatic photogrammetric processes, laser scanning, optical scanners, RGB-D sensors etc.). Primarily, however, the focus is on subsequent post-processing and manipulation of 3D data and their draping with photo-texture. Ultimate goal is to provide students with skills regarding the generation of representations of 3D space with adequate geometric accuracy and high optical quality. Successful completion of this course means that students:

• Have understood and may describe, analyze and compare the different 3D data structures and their characteristics (pros – cons)
• Will be in position to choose and apply the contextually optimal method for collecting 3D information
• Has an in-depth understanding and may apply methods for the mutual registration of point clouds, and also is in position to implement basic registration algorithms.
• Is capable of processing 3D data (smoothing, hole-filling, decimation, normalization) and choose the optimal processing approach according to the type and accuracy of the data involved.
• Has understood and may describe, analyze and compare methods of triangulation and 3D surface generation.
• Is capable of registering image blocks to 3D models and rendering them with photo- texture from the images, by also considering the particular requirements of the representation.
• Is in position to study and evaluate 3D models as regards resolution, accuracy, homogeneity and type of data.
• Can implement an end-to-end process of collection, processing, visualization and analysis of date representing 3D space.

 

General Competences

• Search for, analysis and synthesis of data and information, with the use of the necessary technology
• Working independently
• Teamwork
• Production of free, creative and inductive thinking
• Working in an interdisciplinary environment

 

SYLLABUS

1. Types of 3D data (point cloud, 3D mesh, surfaces, NURBS, Solids).
2. Registration of 3D point clouds (ICP registration, 3D feature matching).
3. 3D data processing (smoothing, hole filling, relaxing, segmentation, classification, object recognition). Triangulation of point sets (2D/3D Delaunay triangulation – Voronoi diagram).
4. Algorithms for surface generation from point clouds, 3D mesh parameterization.
5. DTM generation from DSM.
6. Photogrammetric triangulation sensors (structured light scanners, laser-line scanner, Kinect), laser scanners, optical scanners, combination of photogrammetry and laser scanning.
7. “True” ortho, Photo-texturing of 3D models and their projections (texture mapping, texture atlas, ortho and true ortho, perspective views).
8. 3D city models with level of detail (LOD), 3D mobile mapping (laser-based and image- based).
9. 3D Printing.

 

STUDENT PERFORMANCE EVALUATION

Language of evaluation: Greek
Methods of Evaluation:
• Written examination in the end of the semester (50%), which combines open-ended questions and numeric calculations.
• Evaluation of performance in laboratory exercises (50%)

 

SUGGESTED BIBLIOGRAPHY

1. Course Notes and Slides (in Greek).
2. Botsch M., Kobbelt L., Pauly M., Alliez P., Levy B., 2010. Polygon Mesh Processing. CRC Press
3. Hormann Κ., Levy Β, and Sheffer Α., 2008. Mesh Parameterization: Theory and Practice. Course Notes, SIGGRAPH Asia.