prikaz prve stranice dokumenta Vizualizacija oblaka točaka iz dubinske mape jednog kadra
Access restricted to authenticated users
master's thesis
Vizualizacija oblaka točaka iz dubinske mape jednog kadra
2021. urn:nbn:hr:179:154279
Budimir, Ivo
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Cite this document

Budimir, I. (2021). Vizualizacija oblaka točaka iz dubinske mape jednog kadra (Master's thesis). Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:154279

Budimir, Ivo. "Vizualizacija oblaka točaka iz dubinske mape jednog kadra." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. https://urn.nsk.hr/urn:nbn:hr:179:154279

Budimir, Ivo. "Vizualizacija oblaka točaka iz dubinske mape jednog kadra." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. https://urn.nsk.hr/urn:nbn:hr:179:154279

Budimir, I. (2021). 'Vizualizacija oblaka točaka iz dubinske mape jednog kadra', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 23 March 2023, https://urn.nsk.hr/urn:nbn:hr:179:154279

Budimir I. Vizualizacija oblaka točaka iz dubinske mape jednog kadra [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2021 [cited 2023 March 23] Available at: https://urn.nsk.hr/urn:nbn:hr:179:154279

I. Budimir, "Vizualizacija oblaka točaka iz dubinske mape jednog kadra", Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:179:154279

Please login to the repository to save this object to your list.
Metadata
TitleVizualizacija oblaka točaka iz dubinske mape jednog kadra
Title (english)Visualization of a point cloud from a single shot depth map
AuthorIvo Budimir
MentorMladen Russo (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2021-09-22, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Cilj ovog diplomskog rada je istražiti područje trodimenzionalne rekonstrukcije na temelju postupaka računalnog vida te vizualizirati point cloud-ove dobivene iz depth map-a jednog frame-a snimke snimljene korištenjem time-of-flight i stereo vision kamera. U sklopu rada prikazani su i objašnjeni osnovni pojmovi i koraci procesa 3D rekonstrukcije te je napravljen sustav za vizualizaciju rekonstruiranih point cloud-ova. Provedeno je testiranje kvalitete depth map-a dobivenih time-of-flight senzorom dubine Microsoft Hololens 2 uređaja i stereo vision ZED mini kamere. Testiranja su pokazala da oba navedena uređaja, odnosno senzora dubine imaju svoje prednosti i nedostatke ovisno o karakteristikama površina objekata na sceni koja se snima. Iz vizualizacije rekonstruiranih point cloud-ova može se zaključiti da ToF senzor dubine pokazuje nedostatke kod određivanja dubine refleksivnih, transparentnih i crnih površina, dok nema problema s velikim površinama bez tekstura. S druge strane stereo vision kamera je dobro rekonstruirala refleksivne i transparentne objekte, a pokazala je nedostatke kod izračuna udaljenosti većih površina konstantne boje i bez teksture.
Abstract (english)
The goal of this paper is to explore the field of three-dimensional reconstruction based on computer vision procedures and to visualize point clouds obtained from depth maps of a single frame taken using time-of-flight and stereo vision cameras. This paper explains the basic concepts and steps of 3D reconstruction process, and as a part of the paper the system for visualization of reconstructed point clouds was made. The quality of depth maps obtained by time-of-flight sensor of Microsoft Hololens 2 device and stereo vision ZED mini camera was tested. Tests showed that both of devices, i.e., depth sensors, have their advantages and disadvantages depending on the characteristics of the surfaces of objects in the scene being reconstructed. From the visualization of the reconstructed point-clouds, it can be concluded that ToF depth sensors have limitations in determining the depths of reflective, transparent, and black surfaces, while there are no problems with large areas without textures. On the other hand, the stereo vision camera reconstructed reflective and transparent objects without problems, but it has problems with determining the distance of large areas of constant colour and without texture.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:154279
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2022-07-06 08:17:38