MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model

31 Aug 2022  ·  Mingyuan Zhang, Zhongang Cai, Liang Pan, Fangzhou Hong, Xinying Guo, Lei Yang, Ziwei Liu ·

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

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Motion Synthesis HumanML3D MotionDiffuse FID 0.630 # 27
Diversity 9.410 # 24
Multimodality 1.553 # 21
R Precision Top3 0.782 # 22
Motion Synthesis KIT Motion-Language MotionDiffuse FID 1.954 # 21
R Precision Top3 0.739 # 16
Diversity 11.10 # 3
Multimodality 0.730 # 23

Methods