본문 바로가기
  • Home

A Comparative Study on Character Animation Acting Expression between Generative AI and Keyframe Methods - Focusing on Non-verbal Acting Elements

  • Journal of Software Assessment and Valuation
  • Abbr : JSAV
  • 2025, 21(4), pp.141~148
  • Publisher : Korea Software Assessment and Valuation Society
  • Research Area : Engineering > Computer Science
  • Received : December 1, 2025
  • Accepted : December 20, 2025
  • Published : December 26, 2025

Hunjin Park 1

1중부대학교

Accredited

ABSTRACT

This study examines the limitations of generative AI in character animation's core aspect of 'acting', despite its efficiency improvements. 3D keyframe animation is an art of simplicity and clarity that refines references to convey meaning precisely. In contrast, generative AI relies on probabilistic data combinations, failing to understand narrative context and character intent, producing only patterned gestures and subtext. This paper systematically classifies non-verbal acting elements and compares keyframe animation with AI-generated videos to identify why AI achieves visual realism but lacks acting appeal. It proposes that AI will serve as a production support tool, while final acting direction remains the exclusive domain of human animators.

Citation status

* References for papers published after 2024 are currently being built.