AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Unlocking Accurate Geometry in 3DGS via Spatially Varying Affine Rectification

Tze Ho Elden Tse, Jizong Peng, Richard Chen, Angela Yao

NeurIPS 2026 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes spatially varying affine rectification of monocular depth priors to regularize 3D Gaussian Splatting for more accurate geometry.

Problem

3DGS produces photorealistic novel views but its geometry is inaccurate; existing depth-guided regularization uses monocular depth priors aligned with a single global scale-and-shift, which the authors argue is flawed.

Approach

Replaces the single global scale-and-shift alignment of monocular depth with spatially varying affine rectification, since modern monocular depth estimators are not consistent under one global affine transform.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add L_depth (global scale anchor) to Splatfacto (ablation)Splatfacto · MuSHRoom coffee_roomAbs Rel: 0.059 -> 0.043Improved
add L_grad alone to Splatfacto (ablation)Splatfacto · MuSHRoom coffee_roomF-score: 0.728 -> 0.673Hurt
add L_grad to L_normal + L_depth (ablation)Full method · MuSHRoom coffee_roomF-score: 0.813 -> 0.855Improved
add L_normal to L_grad + L_depth (ablation)Full method · MuSHRoom coffee_roomF-score: 0.850 -> 0.855No clear effect
add all geometric regularizers to Splatfacto (ablation)Splatfacto · MuSHRoom coffee_roomPSNR: 25.293 -> 24.947Hurt
partition scene into K=5 local affine regions instead of one global affine (ablation)Depth-Anything-V2 alignment · MuSHRoominlier ratio (0.02 threshold): 58.9% -> 92.7%Improved
rectify monocular depth with spatially varying affine fields instead of global alignmentDN-Splatter (Monodepth) · MuSHRoom left splitAbs Rel: 0.3542 -> 0.0551Improved
rectify monocular depth instead of global alignment (mesh)DN-Splatter (Monodepth) · MuSHRoom meshF-score: .2765 -> .7097Improved
rectified monocular prior vs sparse-view baselines3DGS · NeRF-LLFF mean, 2-viewPSNR: 17.31 -> 18.79 (vs DRGS)Improved
rectified monocular prior on ScanNet++ meshesSplatfacto · ScanNet++ meshF-score: .3207 vs SuGaR .4367No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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.

Surprising findings

Experimental setup

Builds on: 3D Gaussian Splatting

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