MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model

TL;DR

MotionDiffuse introduces a diffusion model-based framework for text-driven human motion generation, enabling diverse, realistic, and controllable motion synthesis from natural language inputs, surpassing existing methods in quality and flexibility.

Contribution

It is the first diffusion model approach for text-driven human motion generation, offering probabilistic mapping, realistic synthesis, and multi-level manipulation capabilities.

Findings

Outperforms state-of-the-art methods in text-driven motion generation.

Generates diverse and vivid motion sequences.

Provides fine-grained control over body parts and motion length.

Abstract

Human motion modeling is important for many modern graphics applications, which typically require professional skills. In order to remove the skill barriers for laymen, recent motion generation methods can directly generate human motions conditioned on natural languages. However, it remains challenging to achieve diverse and fine-grained motion generation with various text inputs. To address this problem, we propose MotionDiffuse, the first diffusion model-based text-driven motion generation framework, which demonstrates several desired properties over existing methods. 1) Probabilistic Mapping. Instead of a deterministic language-motion mapping, MotionDiffuse generates motions through a series of denoising steps in which variations are injected. 2) Realistic Synthesis. MotionDiffuse excels at modeling complicated data distribution and generating vivid motion sequences. 3) Multi-Level Manipulation. MotionDiffuse responds to fine-grained instructions on body parts, and arbitrary-length motion synthesis with time-varied text prompts. Our experiments show MotionDiffuse outperforms existing SoTA methods by convincing margins on text-driven motion generation and action-conditioned motion generation. A qualitative analysis further demonstrates MotionDiffuse's controllability for comprehensive motion generation. Homepage: https://mingyuan-zhang.github.io/projects/MotionDiffuse.html