Дослідження методів навчання з підкріпленням для створення анімацій рухів персонажів

Abstract

This work is devoted to the issues of animating characters’ movements. Real-time character animation for gaming and film industries is challenging and achieving production-ready quality requires a huge amount of time and resources. Animation through marker-based motion capture is quite a tiresome process that requires costly motion-capture suits, multiple cameras, and a large database. With the use of neural networks, it is possible to generate animation in real-time. We propose using a model that aims to generate real-time human-like character animation in Unity using Reinforcement learning and Deep Reinforcement learning algorithms.