Language-guided Human Motion Synthesis with Atomic Actions

TL;DR

This paper introduces ATOM, a novel approach for language-guided human motion synthesis that decomposes complex motions into atomic actions and uses curriculum learning to improve generalization and realism in generated motions.

Contribution

The paper proposes a new atomic action decomposition method combined with curriculum learning to enhance the diversity and coherence of text-guided human motion synthesis.

Findings

Improved generalization to novel actions.

Enhanced realism and coherence in synthesized motions.

Effective text-to-motion and action-to-motion synthesis results.

Abstract

Language-guided human motion synthesis has been a challenging task due to the inherent complexity and diversity of human behaviors. Previous methods face limitations in generalization to novel actions, often resulting in unrealistic or incoherent motion sequences. In this paper, we propose ATOM (ATomic mOtion Modeling) to mitigate this problem, by decomposing actions into atomic actions, and employing a curriculum learning strategy to learn atomic action composition. First, we disentangle complex human motions into a set of atomic actions during learning, and then assemble novel actions using the learned atomic actions, which offers better adaptability to new actions. Moreover, we introduce a curriculum learning training strategy that leverages masked motion modeling with a gradual increase in the mask ratio, and thus facilitates atomic action assembly. This approach mitigates the overfitting problem commonly encountered in previous methods while enforcing the model to learn better motion representations. We demonstrate the effectiveness of ATOM through extensive experiments, including text-to-motion and action-to-motion synthesis tasks. We further illustrate its superiority in synthesizing plausible and coherent text-guided human motion sequences.