A Survey on Large Language Models for Motion Generation from Text

Obludzyner A.; Zaldivar F.; Ramos O.E. · Proceedings of the 2024 IEEE 31st International Conference on Electronics, Electrical Engineering and Computing, INTERCON 2024 (2024) · DOI: 10.1109/intercon63140.2024.10833491

Resumen

Over these last years, Large Language Models (LLMs) have evolved as an ubiquitous tool for different applications. One of those rapidly growing applications is motion generation, traditionally performed by a specific human design. This paper provides an overview of how motion is generated from text with the use of Large Language Models, targeted at 3D human representations and robots. We describe the different methodologies and techniques of the most used architectures, and the motion generation process in order to achieve the final motion. This survey aims to provide a comprehensive resource for researchers and developers seeking to leverage the power of LLMs to bridge the gap between text understanding and motion generation.

Instituciones: Universidad de Ingeniería y Tecnología