AI Paper Atlas

Research map / 3D vision / Face and shape

NeuFrameQ: Neural Frame Fields for Scalable and Generalizable Anisotropic Quadrangulation

Ying-Tian Liu, Jiajun Li, Yu-Tao Liu, Xin Yu, Yuan-Chen Guo, Yan-Pei Cao, Ding Liang, Ariel Shamir, Song-Hai Zhang

ICCV 2025 · 3D vision › Face and shape · approach: Textured mesh generation · 4 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 polyvector representation instead of raw frame vectors (ablation)frame field regression · DFaustmean angle error: 6.414 -> 6.214Improved
Use polyvector representation instead of raw frame vectors (ablation)frame field regression · Our datasetmean angle error: 11.916 -> 9.231Improved
Use polyvector representation instead of raw frame vectors (ablation)frame field regression · DFaustmean magnitude error: 0.0974 -> 0.0956Improved
Use polyvector representation instead of raw frame vectors (ablation)frame field regression · Our datasetmean magnitude error: 0.5627 -> 0.4247Improved
Add two-stage decomposition (direction regression then diffusion magnitude) (ablation)polyvector regression · DFaustmean angle error: 6.214 -> 5.316Improved
Add two-stage decomposition (direction regression then diffusion magnitude) (ablation)polyvector regression · Our datasetmean angle error: 9.231 -> 6.303Improved
Add two-stage decomposition (direction regression then diffusion magnitude) (ablation)polyvector regression · DFaustmean magnitude error: 0.0956 -> 0.0811Improved
Add two-stage decomposition (direction regression then diffusion magnitude) (ablation)polyvector regression · Our datasetmean magnitude error: 0.4247 -> 0.3289Improved
Use NeuFrameQ frame fields instead of LDFframe field prediction · DFaustmean angle error: 6.987 -> 5.316Improved
Use NeuFrameQ frame fields instead of IntrinsicVHframe field prediction · DFaustmean magnitude error: 0.1043 -> 0.0811Improved
Use NeuFrameQ quadrangulation instead of Instant-Meshesquad remeshing · QuadWild-300 nicolochamfer distance (x1e-5): 0.269 -> 0.757Hurt
Use NeuFrameQ quadrangulation instead of QuadWildquad remeshing · QuadWild-300 lockchamfer distance (x1e-5): 0.587 -> 0.035Improved

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