Prepoznavanje gesti za vizijske senzore zasnovane na događajima pomoću konvolucijskih neuronskih mreža

undergraduate thesis

• Title (english): Gesture recognition for event cameras with convolutional neural networks
• Author: Ante Marić
• Mentor: Ivan Petrović (mentor)
• Committee member: Ivan Petrović (predsjednik povjerenstva)
• Committee member: Antea Hadviger (član povjerenstva)
• Committee member: Luka Petrović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-08-31, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Vizijski senzori zasnovani na događajima biološki su inspirirani senzori koji asinkrono detektiraju promjene u intenzitetu svjetline u sceni neovisno za svaki piksel s vrlo visokom vremenskom razlučivošću. Konvolucijske neuronske mreže sveprisutan su model strojnog učenja u području računalnog vida. Opisan je princip rada vizijskih senzora zasnovanih na događajima te dan pregled metoda koje se koriste u njihovoj primjeni. Nadalje, opisana je implementacija sustava za prepoznavanje geste za ove senzore pomoću konvolucijske neuronske mreže bazirane na grafovima.

Abstract (english)

Event cameras are biologically inspired sensors that asynchronously detect brightness intensity changes in the scene independently for each pixel with a high temporal resolution. Convolutional neural networks are a widely used machine learning model in computer vision. The principles of operation of event cameras are described and an overview of methods used in their application is given. Furthermore, the implementation of a gesture recognition system for these sensors using graph convolutional networks is described.

• Keywords: vizijski senzori zasnovani na događajima; konvolucijske neuronske mreže; prepoznavanje geste; neuromorfičko inžinjerstvo
• Keywords (english): event cameras; convolutional neural networks; gesture recognition; neuromorphic engineering
• Language: english
• URN:NBN: urn:nbn:hr:168:456460
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 09:23:49