Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters

ArXi:2604.05394v1 Announce Type: new Physics-based character animation has become a fundamental approach for synthesizing realistic, physically plausible motions. While current data-driven deep reinforcement learning (DRL) methods can synthesize complex skills, they struggle to reproduce exaggerated, stylized motions, such as instantaneous dashes or mid-air trajectory changes, which are required in animation but violate standard physical laws.