Simulation of a television speaker with natural facial gestures

scientific master's theses

• Title (croatian): Simulacija televizijskog govornika s prirodnim pokretima lica
• Author: Karlo Šmid
• Mentor: Igor Sunday Pandžić (mentor)
• Committee member: Ignac Lovrek (predsjednik povjerenstva)
• Committee member: Igor Sunday Pandžić (član povjerenstva)
• Committee member: Robert Manger (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Telecommunications); Zagreb
• Defense date and country: 2004-11-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

Autonomous Speaker Agent is a graphically embodied animated agent (a virtual character) capable of reading plain English text and rendering it in form of speech accompanied by appropriate, natural-looking facial gestures. The system uses lexical analysis and statistical models of facial gestures in order to generate gestures related to spoken text. It is intended for automatic creation of realistically animated virtual speakers, such as newscasters and storytellers, and incorporates the characteristics of such speakers captured from training video clips. Autonomous Speaker Agent is based on a visual text-to-speech system which generates lip movement synchronized with generated speech. This is extended to include eye blinks, head and eyebrow motion, and a simple gaze following behavior. The result is full face animation produced automatically from plain English text.

Abstract (croatian)

Autonomni razgovorni agent je grafički utjelovljen animirani agent (virtualni lik) koji čita tekst napisan na engleskom jeziku te ga predstavlja kao govor koji je popraćen sa odgovarajućim, prirodnim pokretima lica. Sustav koristi leksičku analizu i statističke modele gesti lica kako bi mogao generirati odgovarajuće geste koje su povezane sa engleskim tekstom. Takav sustav je namijenjen za automatsku generaciju realističnih virtualnih govornika, kao što su TV govornici i pripovjedači, te uključuje karakteristike takvih govornika koje su sakupljene analizom video sekvenci pravih govornika (procesom "treniranja"). Autonomni razgovorni agent je baziran na vizualnom tekst-u-govor sustavu koji generira pokrete usana sinkronizirane sa generiranim govorom. Taj sustav je proširen tako da sada uključuje treptaje očiju, pokrete glave i obrva te jednostavno praćenje pogleda. Rezultat je automatska, potpuna animacija lica na osnovu unesenog engleskog teksta.

• Keywords: Facial animation; virtual character; visual text-to-speech; MPEG-4 FBA; embodied conversational characters
• Keywords (croatian): Animacija lica; virtualni govornik; vizualan tekst-u-govor; MPEG-4 FBA; utjelovljeni razgovorni agenti
• Language: english
• URN:NBN: urn:nbn:hr:168:210221
• Study programme: Title: Postgraduate master programe in computing; Study programme type: university; Study level: postgraduate; Academic / professional title: magistar inženjer računarstva (magistar inženjer računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-07-19 10:16:05