AToM: Aligning Text-to-Motion Model at Event-Level with GPT-4Vision Reward

TL;DR

This paper introduces AToM, a framework that improves text-to-motion alignment at the event level by leveraging GPT-4Vision for detailed annotation and reinforcement learning, resulting in more accurate motion generation from textual prompts.

Contribution

The paper presents a novel approach combining GPT-4Vision-based annotation with reinforcement learning to enhance event-level text-to-motion alignment.

Findings

Significant improvement in event-level alignment quality.

Effective use of GPT-4Vision for detailed motion annotation.

Enhanced motion generation accuracy from textual prompts.

Abstract

Recently, text-to-motion models have opened new possibilities for creating realistic human motion with greater efficiency and flexibility. However, aligning motion generation with event-level textual descriptions presents unique challenges due to the complex relationship between textual prompts and desired motion outcomes. To address this, we introduce AToM, a framework that enhances the alignment between generated motion and text prompts by leveraging reward from GPT-4Vision. AToM comprises three main stages: Firstly, we construct a dataset MotionPrefer that pairs three types of event-level textual prompts with generated motions, which cover the integrity, temporal relationship and frequency of motion. Secondly, we design a paradigm that utilizes GPT-4Vision for detailed motion annotation, including visual data formatting, task-specific instructions and scoring rules for each sub-task. Finally, we fine-tune an existing text-to-motion model using reinforcement learning guided by this paradigm. Experimental results demonstrate that AToM significantly improves the event-level alignment quality of text-to-motion generation.