Sustav za raspoznavanje gesta hrvatskoga znakovnog jezika

master's thesis

• Title (english): System for Recognition of Croatian Sign Language Gestures
• Author: Nikola Tomažin
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Vladimir Čeperić (član povjerenstva)
• Committee member: Zoran Kalafatić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Raspoznavanje gesta znakovnoga jezika veoma je kompleksan problem koji se prvotno temelji na kvalitetnoj procjeni poze. U ovom su radu dane teorijske osnove znakovnoga jezika (u čemu su puno pomogli članovi Hrvatskog saveza gluhih i nagluhih) te strojnog i dubokog učenja te je napravljena usporedba raznih alata procjene poze, od kojih se MediaPipe biblioteka najviše istaknula svojom robusnošcu i radom u stvarnom vremenu. Razvijen je sustav za izgradnju podatkovnoga skupa, treniranje sustava te testiranje preko racunalnoga programa i preko web sučelja.

Abstract (english)

Recognizing sign language gestures is a very complex problem, which is initially
based on a quality pose estimation. This thesis presents the theoretical foundations of
sign language (where members of the Croatian Association of the Deaf and Hard of
Hearing helped a lot), machine and deep learning, and a comparison of various pose
estimation tools, of which MediaPipe stood out for its robustness and real-time work.
A system for building a dataset, system training and testing via a computer program
and via a web interface has been developed.

• Keywords: hrvatski znakovni jezik; racunalni vid; procjena poze; MediaPipe; ˇ gesta; duboko ucenje; povratne neuronske mreže
• Keywords (english): Croatian Sign Language; Computer Vision; Pose Estimation; MediaPipe; Gesture; Deep Learning; Recurrent Neural Networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:570125
• 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 14:01:57