Trodimenzionalna rekonstrukcija čovjeka dubokim modelom

undergraduate thesis

• Title (english): 3D Human Reconstruction Using Deep Learning
• Author: Manuel Matošević
• Mentor: Tomislav Pribanić (mentor)
• Committee member: Tomislav Pribanić (predsjednik povjerenstva)
• Committee member: Kristijan Bartol (član povjerenstva)
• Committee member: David Bojanić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-09-02, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Duboko učenje kao dio strojnog učenja koristi se u brojnim aspektima ljudskoga života.
Automatizirani sustavi za raspoznavanje ljudskoga tijela oslanjanju se na brojne
metode čiji je cilj postiči što višu stopu točnosti pri raspoznavanju, nadilazeći brojne
probleme s kojima se sustavi suočavaju. Pri procesu treniranja pojavili su se raznorazni
problemi koji su značajno skalirali složenost rada. U konačnici, uspješno se poveže 3D
oblik tijela sa 2D slikom, ali se teži k poboljšanju modela. Napredak po tom pitanju
moguće je ostvariti proširivanjem skupa podataka i poboljšavanjem modela (isprobavanje
različitih vrijednosti hiperparametara, drugačijom arhitekturom mreže,...) koje
je isprobano u ovom radu.

Abstract (english)

Deep learning as part of machine learning is used in many aspects of human life.
Automated systems for recognizing the human body rely on a number of methods
aimed at achieving the highest possible rate of recognition accuracy, overcoming a
number of problems faced by the process. During the training process, various problems
arose that significantly scaled the complexity of the work. Ultimately, they
successfully associate the 3D body shape with the 2D image, but also model strives for
improvement. Progress in this regard can be achieved by expanding the data set and
improving the model (testing different values of hyperparameters, different network
architecture, ...) which has been tested in this paper

• Keywords: Duboko učenje; Strojno učenje; Neuronske mreže
• Keywords (english): Deep Learning; Machine Learning; Neural Networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:073895
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 21:09:25