Modeli dubokog učenja za povećanje rezolucije video sekvence u stvarnom vremenu

master's thesis

• Title (english): Deep Learning Models for Real-Time Video Upscaling
• Author: Dino Grgić
• Mentor: Željka Mihajlović (mentor)
• Committee member: Željka Mihajlović (predsjednik povjerenstva)
• Committee member: Jan Šnajder (član povjerenstva)
• Committee member: Alan Jović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-13, Croatia

Abstract

Ključni fokus ovog rada je tehnike za povećanje rezolucije videa i slike u stvarnome vremenu, specifično je naglašena optimizacija dubokih modela kako bi uspjeli postići performanse potrebne za rad u stvarnom vremenu. Rezultati ovog rada demonstriraju sve potencijalne načine kako se CUDA i ONNX mogu iskoristi snagu GPU i optimizacije kako bi se postigao rad u stvarnome vremenu.

Abstract (english)

The primary focus of this thesis revolves around techniques for real-time image and video upscaling, specifically emphasizing the optimization of deep learning models to attain the necessary speed for real-time inference. The results of this thesis provide a comprehensive demonstration of the potential of CUDA and ONNX in harnessing the capabilities of GPUs and optimization methods to achieve real-time image and video inference.

• Keywords: povećanje rezolucije slike; povećanje rezolucije videa; real time; duboko učenje; ONNX
• Keywords (english): image upscale; video upscale; real-time; deep learning; ONNX
• Language: english
• URN:NBN: urn:nbn:hr:168:389809
• 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: 2023-10-11 08:56:57