AI Paper Atlas

Research map / Video / Video generation

FlowerDance: MeanFlow for Efficient and Refined 3D Dance Generation

Kaixing Yang, Xulong Tang, Ziqiao Peng, Xiangyue Zhang, Chubin Chen, xukun zhou, Puwei Wang, Hongyan Liu, Jun He

ECCV 2026 · Video › Video generation · approach: Text-to-motion diffusion · 32 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
replace rectified flow with MeanFlow (20 steps) (ablation)FlowerDance · FineDanceFIDk: 49.59 -> 29.73Improved
reduce MeanFlow steps from 20 to 10 (ablation)FlowerDance · FineDanceFIDk: 29.73 -> 26.17No clear effect
reduce MeanFlow steps from 20 to 5 (ablation)FlowerDance · FineDanceFIDg: 19.59 -> 55.96Hurt
add Physical Consistency Constraint (position+velocity losses) (ablation)MeanFlow · FineDanceFIDk: NaN -> 29.73Improved
add position and velocity losses to reconstruction loss (ablation)PCC · FineDanceFIDk: 37.45 -> 29.73Improved
add velocity loss on top of reconstruction+position (ablation)PCC · FineDanceFIDk: 25.86 -> 29.73Mixed
use BiMamba instead of Mamba (ablation)FlowerDance · FineDanceFIDk: 39.40 -> 29.73Improved
use BiMamba instead of Transformer (NAR) (ablation)FlowerDance · FineDanceall metrics: NaN -> 29.73Improved
use BiMamba instead of inpainting Transformer (II) (ablation)FlowerDance · FineDanceFIDk: 23.29 -> 29.73Mixed
use element-wise addition instead of cross-attention fusion (ablation)FlowerDance · FineDanceFIDk: 32.40 -> 29.73Improved
add time-decayed soft mask in motion editing (ablation)FlowerDance · motion editingboundary continuity (qualitative)Improved
FlowerDance vs MEGA on inference speedFlowerDance · FineDance efficiencyFPS@10^4: 340 -> 4145Improved

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