AI Paper Atlas

Research map / Video / Video generation

PhysDiff: Physics-Guided Human Motion Diffusion Model

Ye Yuan, Jiaming Song, Umar Iqbal, Arash Vahdat, Jan Kautz

ICCV 2023 · Video › Video generation · approach: Text-to-motion diffusion · 313 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add physics-based projection during diffusion sampling to MDMMDM · HumanML3D text-to-motionPhys-Err: 31.572 -> 4.111Improved
add physics-based projection during diffusion sampling to MDMMDM · HumanML3D text-to-motionFID: 0.544 -> 0.433Improved
add physics-based projection to MotionDiffuseMotionDiffuse · HumanML3D text-to-motionPhys-Err: 30.652 -> 2.690Improved
add physics-based projection to MDM on action-to-motionMDM · UESTC action-to-motionFID: FID 12.81 -> 13.27; Phys-Err 28.371 -> 1.463Mixed
increase number of projection steps beyond about four (ablation)PhysDiff · HumanML3D text-to-motionFID / R-PrecisionHurt
increase number of projection steps (ablation)PhysDiff · HumanML3D text-to-motionPhys-ErrImproved
place projection steps consecutively at end (End 4, Space 1) instead of uniform spread (ablation)PhysDiff · HumanML3D text-to-motionPhys-Err: 4.921 -> 4.111Improved
place projection steps at start of diffusion (Start 3, End 1) (ablation)PhysDiff · HumanML3D text-to-motionFID: 0.433 -> 0.510Hurt
reduce spacing between end projections from 3 to 1 (ablation)PhysDiff · HumanML3D text-to-motionPhys-Err: 4.689 -> 4.111Improved
apply physics projection as multi-step post-processing instead of inside diffusion (ablation)MDM · HumanML3D text-to-motionmotion quality / physical plausibilityHurt
add physics simulation to sampling (inference cost)MDM · inference time, single motionseconds: 19.6s -> 51.6sHurt

Later papers that compared with this method

2 later accepted papers report PhysDiff as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy