AI Paper Atlas

Research map / Video / Video generation

ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model

Mingyuan Zhang, Xinying Guo, Liang Pan, Zhongang Cai, Fangzhou Hong, Huirong Li, Lei Yang, Ziwei Liu

ICCV 2023 · Video › Video generation · approach: Text-to-motion diffusion · 241 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 hybrid retrieval and semantics-modulated attention to the diffusion model (ablation)MotionDiffuse-based baseline · KIT-ML testFID: 0.245 -> 0.155Improved
use both semantic and kinematic similarity in retrieval instead of semantic only (ablation)ReMoDiffuse retrieval · KIT-ML testFID: 0.307 -> 0.155Improved
use text features in the attention over retrieved samples (ablation)ReMoDiffuse · KIT-ML testFID: 0.192 -> 0.155Improved
retrieve semantics only (lambda=0 kinematic weight) instead of balancing with lengthhybrid retrieval · KIT-ML and HumanML3Dretrieval FIDHurt
use 2 retrieved samples instead of 1 or 3 (ablation)ReMoDiffuse · KIT-ML testFID: 0.186 (1) / 0.155 (2) / 0.217 (3)Mixed
add retrieval to the baseline for rare prompts (ablation)ReMoDiffuse · KIT-ML testtail 5% MM Dist: 6.173 -> 5.439Improved
apply ReMoDiffuse vs MotionDiffuse on HumanML3Dtext-to-motion diffusion · HumanML3D testFID: 0.630 -> 0.103Improved
use retrieval-augmented model on diversity metricsReMoDiffuse vs MDM/MotionDiffuse · HumanML3D testDiversity: 9.410 (MotionDiffuse) -> 9.018Hurt

Later papers that compared with this method

15 later accepted papers report ReMoDiffuse as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.

7 more not listed.

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