Course detail

Reconstruction and Analysis of 3D Scenes

FSI-QRA Acad. year: 2027/2028 Winter semester

The subject concerns the reconstruction and analysis of 3D scenes that are based on point clouds. This research area is important in the reverse engineering, robotics and autonomous systems. First part of the lectures deals with the data acquisition types and algorithms as registration, edge detection, feature extraction. In laboratory we make the measurement with optical scanner ATOS and other hand scanners. We will process the point cloud with software GOM inspect and Rhinoceros. 
 Arduino Engineering Kit will be used for rover control.
Students prepare their own implementation of some of the algorithm in the last lectures of semester.

Language of instruction

Czech

Number of ECTS credits

6

Learning outcomes of the course unit

Upon successful completion of the course, students will be able to:

  1. Describe and compare selected methods of 3D data acquisition.

  2. Apply PCA, SVD, and ICP algorithms to point cloud processing and registration.

  3. Use the RANSAC algorithm to detect geometric objects in point clouds.

  4. Process data acquired using a 3D scanner.

  5. Create and modify geometric models for further processing.

  6. Design and implement a solution to a practical problem involving real 3D data.

  7. Explain and justify the chosen methods and interpret the obtained results.

Entry knowledge

elementary knowledge of mathematical analysis and algebra (matrix, derivative), computer graphics, the recommended is the knowledge of programming language (C, C++, Pascal, atd) and programm (e.g. Matlab).

Rules for evaluation and completion of the course

Assessment is based on students' active participation in computer-based exercises, continuous monitoring of their work, and evaluation of their ability to understand the methods presented and apply them to practical problems involving real 3D data.

Throughout the semester, students work individually or in groups on assigned practical tasks and projects. Assessment also includes either a final individual project or a series of projects completed during the semester.

The assessment focuses primarily on students' understanding of the methods used, their ability to apply these methods correctly, their individual contribution to solving the assigned tasks, and their ability to explain and interpret the results obtained.

The graded course credit is awarded based on a discussion of the completed tasks and projects, during which students demonstrate their understanding of the applied methods and the results achieved.

Assessment is qualitative, without a point-based grading system. If the requirements are not sufficiently met, students are given the opportunity to complete or revise the assigned tasks or projects and have them reassessed.

Aims

The main gist of the subject is to understand the point cloud and its processing algorithms. The practical part will show the 3D scanning and 3D printing technology and the 3D scene algorithms.
Students will learn about the point clouds, its acquisition (3D scanning), usage (3D printing, Arduino scene analysis) and post-processing (edge or object detection, registration).

The study programmes with the given course

Programme N-AAE-P: Advanced Automotive Engineering, Master's, compulsory-optional

Type of course unit

 

Lecture

26 hours, optionally

Syllabus

Lectures:
1.The data acquisition (Terrain, Mobile, Airborne) in dependance of the application. Passive (Structure from Motion) and active methods (Time of Flight, laser)
2. RANSAC – algorithm and its usage, feature extraction
3.-4. Point cloud registration (methods PCA, SVD, ICP, FPFH)
5. Ground point detection, clustering
6. Comparison spatial methods with image analysis methods
7. Laboratory measurements on the optical scanner ATOS and hand scanners
8. Point cloud processing in software (GOM Inspect, Rhinoceros, etc.), work with real scanned data
9. 3D printing – principles, settings, problems
10. Arduino Engineering Kit – vehicle Rover can navigate between given reference points, move objects with a forklift
11.-12. Consultations
13. Presentation of the seminar work

Computer-assisted exercise

26 hours, compulsory

Syllabus

Exercise (with computers):


1. Data capture methods (Terrain, Mobile, Airborne) and their use, SFM Matlab programming, data capture using Intel Real Sense, IPad Pro
2. RANSAC – line, plane, testing
3. -4.: Programming registration methods using predefined functions, influence of parameters on results
5. Land point detection, clustering
6. Image processing methods and their comparison with methods used for 3D (derivative, edge detectors, etc.)
7. Scanning in the laboratory – ATOS
8. Practical exercises with acquired data
9. 3D printing – model design in Rhinoceros, printing
10. Arduino Engineering Kit – programming the Rover and its orientation in space and functions, simple circuit design with Arduino
11. -12. Semester project consultation
13. Presentation of final projects