Investigation of Machine Learning Techniques for Motion Planning and Control in Animation

Abstract

Motion planning and control are crucial components that must be present in animation software if computer-generated characters are to move in a way that appears natural and realistic. Traditional animation techniques usually make use of manual keyframing, which can be labour- and time-intensive depending on the circumstances. As a direct result of these advancements, there has been a rise in the amount of interest shown in investigating the opportunities presented by machine learning (ML) approaches for the purpose of motion planning and control in animation. This research investigates a variety of machine learning (ML) strategies and their applications in motion planning and control. Particular attention is paid to the methodologies' respective benefits, limitations, and challenges. The results of this research provide valuable information that sheds light on the potential for ML to contribute to the development of animation techniques.