Graduate study programme

Back   Schedule   Hrvatski

Computational Geometry and Robot Vision DRb1-18

ECTS 5 | P 30 | A 0 | L 30 | K 0 | ISVU 190630 | Academic year: 2019./2020.

Course groups

Prikaži sve grupe na predmetu

Course lecturers



Explain the representation of basic geometric structures and spatial relations by appropriate data structures. Explain the basic methods for an efficient analysis of image and 3D sensor data. Explain the use of programing tools for image and 3D sensor data processing. Explain how to implement programme solutions to image and 3D sensor data analysis problems with the application in robotics and intelligent autonomous systems.

Conditions for enrollment

Requirements for the enrolment in the graduate university study programme Computer Engineering.

Course description

Basic concepts - coordinate system, point, line, line segment, vector, plane, face, polygon, polyhedron, normal. Description of the position and orientation of a rigid body. Transformation between coordinate systems. Plane and space partition. Triangulation. Delaunay triangulation. Nearest neighbour search. KD-tree. Convex hull. Voronoi diagram. Maps. Point cloud registration. Hough transform. Random Sample Consensus (RANSAC). Application of computer vision in robotics. Perception sensors - camera, 3D camera, stereo vision, LIDAR. Image filtering. Edge and key point detection. Image and 3D point cloud segmentation. Optical flow. Camera calibration. Estimating camera pose relative to the operating environment of a robot. Multiple view reconstruction of a three-dimensional object and scene. Map building using computer vision. Place recognition. Obstacle detection.

Teaching methods

lectures, individual exercises, laboratory exercises

Student requirements

Defined by the Student evaluation criteria of the Faculty of Electrical Engineering, Computer Science and Information Technology Osijek and Paragraph 1.9.

Monitoring of students

Defined by the Student evaluation criteria of the Faculty of Electrical Engineering, Computer Science and Information Technology Osijek and Paragraph 1.9.

Obligatory literature

1. 1 Bradski, G.; Kaehler, A. Learning OpenCV O Reilly, 2008

Pretraži literaturu na:

Recommended additional literature

1. 1 E. R. Davies Machine Vision: Theory, Algorithms, Practicalities, 3rd edition Elsevier, San Francisco, USA, 2005

2. 2 R. Hartley, A. Zisserman Multiple View Geometry in Computer Vision Cambridge University Press, 2003.

3. 3 O. Faugeras Three-Dimensional Computer Vision: A Geometric Viewpoint Cambridge, Massachusetts: The MIT Press, 1993.

4. 4 R. Cupec Osnove inteligentnih robotskih sustava, udžbenik u izradi Zavod za računalno inženjerstvo i automatiku, ETF Osijek, 2014.

Course assessment

Conducting university questionnaires about teachers (teacher-student relationship, criteria transparency, motivation for task completion, clarity of teaching, etc.). Conducting faculty questionnaires about course quality upon their completion (students’ self-assessment on learning outcomes and workload with respect to ECTS credits and overall course activities).

Overview of course assesment

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

1. understand the basic principles of common methods and tools for image and 3D sensor data processing

2. select appropriate methods for solving problems in the field of image and 3D sensor data processing

3. select appropriate data structures for representation of 2D and 3D geometric structures and their relations

4. develop computer programmes for image processing using available software development tools

5. develop computer programmes for 3D sensor data processing using available software development tools

6. develop programme solutions for recognition of objects in 3D point clouds obtained by 3D sensors and estimation of object pose with respect to the camera

Aktivnosti studenta: Vidi tablicu aktivnosti