diplomsko delo
Kristjan Žarn (Author), Danijel Skočaj (Mentor)

Abstract

Za gradnjo 3D modelov so tipično uporabljeni globinski senzorji ali stereo sistemi kamer. Takšne rešitve so manj dostopne, zato je zanimiv izziv na področju računalniškega vida gradnja 3D modelov brez dodatne opreme. V diplomskem delu obravnavamo rekonstrukcijo predmetov na podlagi fotografij. Najprej pregledamo teoretično ozadje in uporabljene algoritme, nato pa predstavimo naš sistem za rekonstrukcijo. S postopkom imenovanim struktura iz gibanja, pridobimo redko rekonstrukcijo predmeta in zunanje parametre kamer. Slednje uporabimo pri gradnji gostega 3D modela. Evalvacija zunanjih parametrov pokaže, da pri dodajanju kamer v sistem pride do akumulacije napake. Povprečna napaka lege kljub temu znaša le nekaj milimetrov. Na koncu ovrednotimo še natančnost pridobljenih modelov tako, da jih primerjamo z referenčnimi. Rezultati pokažejo, da je povprečna napaka rekonstrukcij majhna, zato lahko povzamemo, da implementirani sistem deluje dovolj dobro.

Keywords

struktura iz gibanja;kalibracija kamere;redka rekonstrukcija;gosta rekonstrukcija;3D model;računalništvo;računalništvo in informatika;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [K. Žarn]
UDC: 004.932(043.2)
COBISS: 1536519619 Link will open in a new window
Views: 890
Downloads: 260
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: Building a 3D model with a moving monocular camera
Secondary abstract: Reconstruction of a 3D model is typically obtained using depth sensors or a stereo camera system. Such solutions are less accessible, therefore it is an interesting challenge in the field of computer vision to build a 3D model without additional equipment besides a camera. Our work discusses reconstruction based on images. First we examine the theoretical background and algorithms, then we proceed with description of our reconstruction system. We use a method called structure from motion to obtain object's sparse reconstruction and camera's extrinsic parameters. The latter are used for building a dense 3D model. Evaluation of extrinsic parameters shows that adding cameras to the reconstruction leads to accumulation of the error. Despite this occurrence the average error of the calculated poses remains within a few millimetres. Finally we evaluate the accuracy of the reconstructed models by comparing them with their ground truth. Results indicate small average errors, therefore we can conclude that the implemented system works sufficiently well.
Secondary keywords: structure from motion;camera calibration;sparse reconstruction;dense reconstruction;3D model;computer science;computer and information science;diploma;
File type: application/pdf
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 58 str.
ID: 8966482