Acquisition of basic knowledge in computer vision, 2D projective planar geometry, multi-camera calibration and 3D reconstruction, and 3D visualization.

Obtaining knowledge in basics of computer vision and multi-view camera calibration and 3D reconstruction. 2D planar projective geometry, geometric transformations of images and video. Generating panorama/mozaic image based on multiple acquired images. Knowledge of various methods for image feature detection and description as well as their matching for objection registration, multi-camera camera calibration and 3D reconstruction. Automatic generation of 3D model and its 3D visualization based on the reconstructed 3D cloud and camera image textures.

Geometric transformation of images and video. 2D projective geometry in images and video. 2D feature detection and description in images and video for computer vision tasks. Feature matching and correspondence between multiple images in multi-camera system based on which 2D image alignment is done. Image mozaic/panorama generation for multiple neighboring camera images. Epipolar geometry. 3D reconstruction based on multiple images in multi-camera system. Generating polygon mesh based on reconstructed 3D point cloud. Shading, lighting and final 3D model rendering.

Teaching includes the following forms: theoretical lectures, practical computational classes, practical project assignments and consultation. Lectures include teaching of theoretical concepts, where active participation of students is highly stimulated through discussions. Practical classes are done in computer lab on PCs using C/C++ programming language with supporting computer vision and computer graphic open source libraries: OpenCV, PCL and OpenGL. In computational classes, video cameras are also used where, through specially developed code, multi-camera acquisition is done.