Reverzibilna deidentifikacija u video zapisima

master's thesis

• Title (english): Reversible de-identification in video sequences
• Author: Martin Blažević
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Zoran Kalafatić (član povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Deidentifikacija je proces, odnosno tehnika kojom se podaci o osobi mijenjaju na način da se ne mogu povezati s njezinim identitetom, a ovim je radom pružen teorijski opis i programska implementacija jednog mogućeg rješenja problema deidentifikacije osoba u video zapisima. Objašnjeni su i opisani osnovni principi proširene stvarnosti i stvaranja modela virtualnog čovjeka, te alati korišteni za implementaciju programskog dijela. Microsoft Kinect SDK olakšava razvoj aplikacija koje u radu koriste Kinect dubinsku kameru, dok Microsoft XNA Framework 4.0 radni okvir i XNA Game Studio omogućuju rješavanje zahtjevnih zadataka, poput animacije virtualnih likova u 3D prostoru. Identitet osoba u video zapisima prikriva se iscrtavanjem virtualnih 3D likova na mjestima gdje se osobe pojavljuju u videu, a kako je izvornu informaciju potrebno sačuvati, koriste se tehnike steganografije.

Abstract (english)

De-identification is a process, or better, a technic used to prevent a person's identity from being connected with information. This paper describes one possible solution of de-identification problems in videos. It explains main principles of augmented reality and 3D modeling, and also, tools used for computer programming. Microsoft Kinect SDK simplifies the development of application software which uses Kinect depth sensor, while Microsoft XNA Framework 4.0 framework and XNA Game Studio enables solving difficult tasks, like animation of 3D models. Person's identity in videos is covered up by drawing 3D models in places, coordinates where person appears in videos, and the steganography techniques are used for keeping the original information.

• Keywords: Reverzibilna deidentifikacija; proširena stvarnost; 3D model; Kinect; MakeHuman; animacija; praćenje pokreta; steganografija; XNA; C#; Avateering
• Keywords (english): Reversible de-identification; augmented reality; 3D model; Kinect; MakeHuman; animation; motion tracking; steganography; XNA; C#; Avateering
• Language: croatian
• URN:NBN: urn:nbn:hr:168:840909
• 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: 2018-02-06 07:56:12