AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

MeshFlow: Efficient Artistic Mesh Generation via MeshVAE and Flow-based Diffusion Transformer

Weiyu Li, Antoine Toisoul, Tom Monnier, Roman Shapovalov, Rakesh Ranjan, Ping Tan, Andrea Vedaldi

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 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
use TokenMerge queries instead of Q-Former learnable queries (ablation)MeshVAE · MeshVAE ablationF1 score: 49.47 -> 99.78Improved
use TokenMerge queries instead of FPS queries (ablation)MeshVAE · MeshVAE ablationF1 score: 60.18 -> 99.78Improved
use TokenMerge queries instead of Q-Former (ablation)MeshVAE · MeshVAE ablationVertex distance (x100): 23.36 -> 0.75Improved
use TokenMerge queries instead of FPS (ablation)MeshVAE · MeshVAE ablationVertex distance (x100): 18.29 -> 0.75Improved
increase downsampling ratio from 2 to 4 (ablation)MeshVAE · MeshVAE ablationF1 score: 92.65 -> 88.82Hurt
increase maximum vertex count from 2048 to 8192 (ablation)MeshVAE · MeshVAE ablationF1 score: 99.84 -> 99.71No clear effect
replace AR mesh generation with rectified flow on MeshVAE latentsMeshFlow vs FastMesh-V4K · Toys4K shape-conditioned generationChamfer distance (x100): 4.05 -> 2.33Improved
replace AR mesh generation with rectified flow on MeshVAE latentsMeshFlow vs FastMesh-V4K · Toys4K shape-conditioned generationHausdorff distance (x100): 10.22 -> 4.23Improved
use continuous MeshVAE latents instead of EdgeRunner ArAE tokensMeshVAE · MeshVAE reconstructionCompression ratio: 0.47 -> 0.014Improved
use continuous MeshVAE instead of EdgeRunner on reconstructionMeshVAE · MeshVAE reconstructionChamfer distance (x100): 1.21 -> 1.29No clear effect
use rectified flow instead of FastMesh-V1K autoregression for single-mesh inferenceMeshFlow · Single-mesh inferenceInference time (s): ~21 s -> ~1.2 sImproved

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