Названия:
ADAPTIVE CHARACTER ANIMATION IN GAMES USING MOTION CAPTURE AND REINFORCEMENT LEARNING
Автор:
Rahman Myradov, Sapartach Hojabalkanova, Selim Myradov, Emir Sopyyev
Расположение страниц:
53-58
Язык:
Английский
Аннотация:
In recent years, video games have increasingly demanded realistic and adaptive character animations that can change depending on player actions and game conditions. Traditional animation methods, such as motion capture (MoCap), provide high-quality animations, but are limited in their ability to adapt to dynamic gaming situations. This paper presents a system that integrates MoCap data with reinforcement learning (RL) techniques to create adaptive animations that respond to real-time changes in the game environment and player interactions. We propose a methodology where RL agents are trained to optimize character animations based on the game context. Experimental results demonstrate that this system significantly enhances interactivity and realism, improving the overall gaming experience. In conclusion, we explore potential avenues for future development to increase the versatility and performance of the proposed algorithms.