The influence of sound effects on the perceived smoothness of rendered animations

  • Authors:
  • Georgia Mastoropoulou;Kurt Debattista;Alan Chalmers;Tom Troscianko

  • Affiliations:
  • University of Bristol, UK;University of Bristol, UK;University of Bristol, UK;University of Bristol, UK

  • Venue:
  • APGV '05 Proceedings of the 2nd symposium on Applied perception in graphics and visualization
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The developers and users of interactive computer graphics (CG), such as 3D games and virtual reality, are demanding ever more realistic computer generated imagery delivered at high frame rates, to enable a greater perceptual experience for the user. As more computational power and/or transmission bandwidth are not always available, special techniques are applied that trade off fidelity in order to reduce computational complexity, while trying to minimise the perceptibility of the resulting visual defects. Research on human visual perception has promoted the development of perception driven CG techniques, where knowledge of the human visual system and its weaknesses are exploited when rendering/displaying 3D graphics. It is well known in the human perception community that many factors, including audio stimuli, may influence the amount of cognitive resources available to perform a visual task. In this paper we investigate the influence sound effects have on the perceptibility of motion smoothness in an animation (i.e. on the perception of delivered frame rate). Forty participants viewed pairs of computer-generated walkthrough animations (with the same visual content within the pair) displayed at five different frame rates, in all possible combinations. Both walkthroughs in each test pair were either silent or accompanied by sound effects and the participant had to detect the one that had a smoother motion (i.e. was delivered at higher frame rate). A significant effect of sound effects on the perceived smoothness was revealed. The participants who watched the audiovisual walkthroughs gave more erroneous answers while performing their task compared to the subjects in the "No Sound" group, regardless of their familiarity with animated CG. Especially the unfamiliar participants failed to notice motion smoothness variations which were apparent to them in the absence of sound. The effect of the type of camera movement in the scene (translation or rotation) on the viewers' perception of the motion smoothness/jerkiness was also investigated, but no significant association between them was found. Our results should lead to new insights in 3D graphics regarding the requirements for the delivered frame rate in a wide range of applications.