Free-T2M: Frequency Enhanced Text-to-Motion Diffusion Model With Consistency Loss

Benchmark Model Rank Results
motion-synthesis-on-kit-motion-languageFree-T2M (StableMoFusion)#4FID: 0.155R Precision Top3: 0.789Diversity: 10.902
motion-synthesis-on-humanml3dFree-T2M (StableMoFusion)#7FID: 0.051R Precision Top3: 0.803Diversity: 9.480