FreeMotion: A Unified Framework for Number-free Text-to-Motion Synthesis

TL;DR

FreeMotion introduces a unified, number-free framework for synthesizing both single and multi-person human motions from text, enabling precise control and broad applicability in computer vision tasks.

Contribution

The paper proposes a novel unified framework that combines single and multi-person motion synthesis without requiring explicit person counts, advancing the universality of text-to-motion generation.

Findings

Outperforms existing methods in multi-person motion synthesis

Supports seamless integration of spatial control for multi-person motions

Capable of inferring both single and multi-human motions simultaneously

Abstract

Text-to-motion synthesis is a crucial task in computer vision. Existing methods are limited in their universality, as they are tailored for single-person or two-person scenarios and can not be applied to generate motions for more individuals. To achieve the number-free motion synthesis, this paper reconsiders motion generation and proposes to unify the single and multi-person motion by the conditional motion distribution. Furthermore, a generation module and an interaction module are designed for our FreeMotion framework to decouple the process of conditional motion generation and finally support the number-free motion synthesis. Besides, based on our framework, the current single-person motion spatial control method could be seamlessly integrated, achieving precise control of multi-person motion. Extensive experiments demonstrate the superior performance of our method and our capability to infer single and multi-human motions simultaneously.