AI Paper Atlas

Research map / Video / Video generation

ClusterStyle: Modeling Intra-Style Diversity with Prototypical Clustering for Stylized Motion Generation

Kerui Chen, Jianrong Zhang, Ming Li, Zhonglong Zheng, Hehe Fan

ECCV 2026 · Video › Video generation · approach: Text-to-motion diffusion · 1 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 multi-prototype clustering style encoder instead of learnable classification head with cross-entropy (ablation)ClusterStyle · Stylized motion generationSRA: 3.32% drop in SRA when replacedImproved
add L_intra to L_inter (ablation)ClusterStyle style encoder · HumanML3D+100STYLE stylized motion generationSRA: 75.182 -> 78.101Improved
add L_inter to L_intra (ablation)ClusterStyle style encoder · HumanML3D+100STYLE stylized motion generationSRA: 67.746 -> 78.101Improved
use prototype contrastive L_intra instead of cross-entropy on learnable prototypes (ablation)ClusterStyle style encoder · HumanML3D+100STYLE stylized motion generationSRA: 65.465 -> 78.101Improved
add L_intra to L_inter (ablation)ClusterStyle style encoder · HumanML3D+100STYLE stylized motion generationFID: 1.112 -> 1.137Hurt
use cosine similarity instead of L2 (ablation)ClusterStyle · HumanML3D+100STYLE stylized motion generationSRA: 76.483 -> 78.101Improved
use cosine similarity instead of L1 (ablation)ClusterStyle · HumanML3D+100STYLE stylized motion generationFID: 1.322 -> 1.137Improved
increase number of global prototypes K_g (ablation)ClusterStyle · HumanML3D+100STYLE stylized motion generationR-PrecisionMixed
increase number of local prototypes K_l from 30 to 50 (ablation)ClusterStyle · HumanML3D+100STYLE stylized motion generationSRANo clear effect
use ClusterStyle instead of StyleMotifStyleMotif baseline · Motion style transferFID: 1.375 -> 0.768Improved
use ClusterStyle instead of BiFlowBiFlow baseline · HumanML3D+100STYLE stylized motion generationR-Precision (Top-3): 0.613 -> 0.708Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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