Metoda za dinamičko pozicioniranje objekta u videozapisu

master's thesis

• Title (english): Method for Dynamic Object Positioning in Video Recording
• Author: Marko Bago
• Mentor: Sven Lončarić (mentor)
• Committee member: Sven Lončarić (predsjednik povjerenstva)
• Committee member: Marko Subašić (član povjerenstva)
• Committee member: Tomislav Petković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-09-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

S obzirom na sve veću zasićenost potrošača oglasima, dinamičko umetanje istih u
videozapise ima veliki tržišni potencijal. Suvremena rješenja dinamičkog umetanja sadržaja
u okolini sa „zelenim platnom“ koriste hardverska rješenja poput više kamera na različitim
pozicijama i dubinama te inovativne metode s takozvanim „3-point“ osvjetljenjem za
precizno prepoznavanje rubova predmeta i ljudi u sceni. Zbog toga je izazov postići jednako
uvjerljiv učinak koristeći samo običnu kameru i računalo. U ovome radu predložena je
metoda umetanja slikovnog marketinškog sadržaja u videozapise na zeleno platno sa
markerima. Markeri omogućavaju da se točno aproksimiraju rubne točke platna u
slučajevima kada postoji okluzija, a boja platna nam omogućuje da segmentiramo dio platna
koji nije zaklonjen i na njega ubacimo željenu reklamu.

Abstract (english)

While the advertisement market is constantly growing, a method for dynamic
advertisement placement in videos has a great market potential. Current solutions for
dynamic placement of content using a „green screen“ use expensive hardware solutions like
multiple positional cameras with different depth of field and innovative methods such as „3-
point“ lighting for precise edge recognition of objects and people in the scene. For these
reasons, it is a challenge to get the same quality of results using only a simple camera and a
computer. In this work we present a method for dynamic placing advertisements in video
content using a green canvas („green screen“) with markers. The markers are used for
correctly approximating the corners of the canvas in cases where it is occluded and the color
of the canvas enables to segment its visible part to place the advertisement on it.

• Keywords: Računalni vid; Obrada slike; Proširena stvarnost; Segmentacija slike
• Keywords (english): Computer Vision; Image Processing; Augmented Reality; Image Segmentation
• Language: croatian
• URN:NBN: urn:nbn:hr:168:747872
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 11:42:40