T2M-X: Learning Expressive Text-to-Motion Generation from Partially Annotated Data

TL;DR

T2M-X is a novel two-stage approach that generates expressive, whole-body humanoid animations from text prompts by learning from partially annotated data, improving motion quality and coordination.

Contribution

It introduces a multi-part VQ-VAE and GPT-based framework for comprehensive text-to-motion generation from limited annotations, addressing dataset inconsistency issues.

Findings

Significant quantitative improvements over baselines.

High-quality, expressive whole-body motion outputs.

Robustness against dataset limitations.

Abstract

The generation of humanoid animation from text prompts can profoundly impact animation production and AR/VR experiences. However, existing methods only generate body motion data, excluding facial expressions and hand movements. This limitation, primarily due to a lack of a comprehensive whole-body motion dataset, inhibits their readiness for production use. Recent attempts to create such a dataset have resulted in either motion inconsistency among different body parts in the artificially augmented data or lower quality in the data extracted from RGB videos. In this work, we propose T2M-X, a two-stage method that learns expressive text-to-motion generation from partially annotated data. T2M-X trains three separate Vector Quantized Variational AutoEncoders (VQ-VAEs) for body, hand, and face on respective high-quality data sources to ensure high-quality motion outputs, and a Multi-indexing Generative Pretrained Transformer (GPT) model with motion consistency loss for motion generation and coordination among different body parts. Our results show significant improvements over the baselines both quantitatively and qualitatively, demonstrating its robustness against the dataset limitations.