magistrsko delo
Luka Loboda (Author), Luka Čehovin (Mentor)

Abstract

Določanje lege kamere v obogateni resničnosti zgolj na podlagi vizualne informacije je velik izziv. Problematična je predvsem inicializacija algoritma ob popolni odsotnosti predhodnih informacij o strukturi scene. V tem delu smo razvili rešitev na osnovi obstoječega algoritma za lokalizacijo kamere v obogateni resničnosti (PTAM), ki omogoča vstavitev predhodno rekonstruirane scene ali njenega dela v algoritem za sledenje in s tem omogoči robustno samodejno inicializacijo, posredno pa omogoča tudi določitev skale scene. Metoda sama prepozna trenutno sceno na vhodnem posnetku in uporabi ustrezen model za lokalizacijo. Predlagano rešitev smo evalvirali na testni zbirki posnetkov in rekonstrukcij scen, ki smo jo zajeli s tem namenom. Pokazali smo, da je naš pristop izboljšal uspešnost in natančnost izvornega algoritma PTAM. Našo rešitev smo primerjali tudi z referenčnim algoritmom za sledenje in kartiranje ORB-SLAM2 in pokazali, da naša rešitev dosega večjo uspešnost sledenja ob primerljivi napaki, poleg tega pa deluje hitreje. Na koncu smo s testno aplikacijo prikazali uporabnost naše programske rešitve v obogateni resničnosti.

Keywords

sočasna lokalizacija in kartiranje;samodejna inicializacija;rekonstrukcija scene;obogatena resničnost;računalništvo;računalništvo in informatika;magisteriji;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [L. Loboda]
UDC: 004(043.2)
COBISS: 37287427 Link will open in a new window
Views: 820
Downloads: 209
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Utilizing 3D models to initialize algorithm for camera localization in augmented reality
Secondary abstract: Determining the position of the camera in augmented reality based only on visual information still poses a great challenge. Particularly problematic is initialization of the algorithm in a complete absence of pre-existing information about the structure of the scene. In this thesis, we developed a solution based on an existing algorithm for camera localization in augmented reality (PTAM) which is able to robustly and automatically initialize itself by importing a previously reconstructed scene or part of it into the algorithm for tracking and it can also indirectly determine the scale of the scene. Method recognises the current scene on the input recording and uses appropriate model for localization. The proposed solution was evaluated on a test dataset with a collection of prepared recordings and scene reconstructions. It has been shown that our approach improved the success rate and precision of the original PTAM algorithm. Our solution was also compared with a reference algorithm for tracking and mapping ORB-SLAM2 and it has been shown that our solution achieved a better success rate with comparable error while also being faster. At the end, the usability of our software solution in augmented reality was demonstrated with an example application.
Secondary keywords: simultaneous localization and mapping;automatic initialization;scene reconstruction;augmented reality;computer science;computer and information science;master's degree;
Type (COBISS): Master's thesis/paper
Study programme: 1000471
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 87 str.
ID: 12133083