Crowd-Sourced Procedural Animation Optimisation: Comparing Desktop and VR Behaviour
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review
Standard Standard
2017 International Conference on Cyberworlds (CW). Chester, UK, 2017. p. 48-55.
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review
HarvardHarvard
APA
CBE
MLA
VancouverVancouver
Author
RIS
TY - GEN
T1 - Crowd-Sourced Procedural Animation Optimisation: Comparing Desktop and VR Behaviour
AU - Henshall, Gareth
AU - Ap Cenydd, Llyr
AU - Teahan, William
PY - 2017
Y1 - 2017
N2 - Procedural animation systems are capable of synthesising life-like organic motion automatically. However due to extensive parameterisation, tuning these systems can be very difficult. Not only are there potentially hundreds of interlinked parameters, the resultant animation can be very subjective andthe process difficult to automate effectively. In this paper we describe a crowd-sourced approach to procedural animation parameter optimisation using genetic algorithms. We test our approach by asking users to interactively rate population of virtual dolphins to a prescribed behavioural criteria. Our results show that within a few generations a group of users can successfully tune the system toward a desired behaviour. Our secondary motivation is to investigate if there are differences in animation and behavioural preference between observing on a standard desktop monitor and Virtual Reality (VR). We describe a study where users tuned two sets of dolphin animation systems in parallel, one using a normal monitor and another in VR. Our results indicate that being immersed in VR leads to some key differences in optimised behaviour.
AB - Procedural animation systems are capable of synthesising life-like organic motion automatically. However due to extensive parameterisation, tuning these systems can be very difficult. Not only are there potentially hundreds of interlinked parameters, the resultant animation can be very subjective andthe process difficult to automate effectively. In this paper we describe a crowd-sourced approach to procedural animation parameter optimisation using genetic algorithms. We test our approach by asking users to interactively rate population of virtual dolphins to a prescribed behavioural criteria. Our results show that within a few generations a group of users can successfully tune the system toward a desired behaviour. Our secondary motivation is to investigate if there are differences in animation and behavioural preference between observing on a standard desktop monitor and Virtual Reality (VR). We describe a study where users tuned two sets of dolphin animation systems in parallel, one using a normal monitor and another in VR. Our results indicate that being immersed in VR leads to some key differences in optimised behaviour.
U2 - 10.1109/CW.2017.52
DO - 10.1109/CW.2017.52
M3 - Conference contribution
SN - 978-1-5386-2090-8
SP - 48
EP - 55
BT - 2017 International Conference on Cyberworlds (CW)
CY - Chester, UK
ER -