AI Paper Atlas

Research map / 3D vision / Face and shape

VRP-UDF: Toward Unbiased Learning of Unsigned Distance Functions From Multi-View Images With Volume Rendering Priors

Wen Zhang, Chunsheng Wang, Kanle Shi, Yu-Shen Liu, Zhizhong Han

TPAMI 2026 · 3D vision › Face and shape · approach: Neural implicit surface reconstruction

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-resolution windows (ablation)VRPrior (NeuS backbone) · DTUChamfer distance: 0.85 -> 0.82Improved
add multi-resolution windows (ablation)VRPrior (NeuS backbone) · DF3DChamfer distance: 1.71 -> 1.67Improved
add point sampling priors (ablation)VRPrior with multi-resolution windows · DF3DChamfer distance: 1.67 -> 1.62Improved
add uniform sampling (ablation)VRPrior with windows and point sampling priors · DF3DChamfer distance: 1.62 -> 1.59Improved
use multi-resolution windows {10,20,30} instead of single size 1 (ablation)volume rendering prior · DF3DChamfer distance: 2.26 -> 1.59Improved
use window set {1,30,50} style large span instead of {10,20,30} (ablation)volume rendering prior · DTUChamfer distance: 0.80 -> 0.92Hurt
add sampling interval to prior input (vs only UDF) (ablation)volume rendering prior · DF3DCD-L1: 1.81 -> 1.59Improved
use interval rather than coordinates as prior input (ablation)volume rendering prior · DF3DCD-L1: 2.93 -> 1.59Improved
train prior with depth supervision rather than RGBD (ablation)volume rendering prior · DF3DCD-L1: 2.05 -> 1.59Improved
fine-tune prior during inference (ablation)volume rendering prior · DF3DCD-L1: 1.59 -> does not convergeHurt
apply prior to NeuS SDF rendererNeuS · DTUChamfer distance: 0.98 -> 0.87Improved
apply prior to UNISURF occupancyUnisurf · DTUChamfer distance: 1.22 -> 1.01Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 6 of 7 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