AI Paper Atlas

Research map / 3D vision / 3D generation

PlaneRAS: Learning Planar Primitives for 3D Plane Recovery

Fang Zhang, Wenzhao Zheng, Linqing Zhao, Zelan Zhu, Jiwen Lu, Xiuzhuang Zhou

ICCV 2025 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 0 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
add pretrained monocular depth prior (ablation)PlaneRAS (ResNet50) · ScanNet valVI: 0.701 -> 0.610Improved
add pretrained depth prior (ablation)PlaneRAS (ResNet50) · ScanNet valplane offset error (mm): 161.56 -> 106.61Improved
add normal prior on top of depth prior (ablation)PlaneRAS (ResNet50) · ScanNet valVI: 0.610 -> 0.608No clear effect
add normal prior on top of depth prior (ablation)PlaneRAS (ResNet50) · ScanNet valnormal error (deg): 7.77 -> 7.88No clear effect
use plane kernel instead of 2D Gaussian kernel (no prior) (ablation)PlaneRAS (ResNet50) · ScanNet valVI: 0.762 -> 0.701Improved
use plane kernel instead of 2D Gaussian kernel (no prior) (ablation)PlaneRAS (ResNet50) · ScanNet valSC: 0.840 -> 0.856Improved
reduce primitive resolution from 48x64 to 36x48 (ablation)PlaneRAS (ResNet50) · ScanNet valVI: 0.610 -> 0.630Hurt
reduce primitive resolution from 48x64 to 24x32 (ablation)PlaneRAS (ResNet50) · ScanNet valVI: 0.610 -> 0.653Hurt
reduce primitive resolution from 48x64 to 36x48 (ablation)PlaneRAS (ResNet50) · ScanNet valnormal error (deg): 7.77 -> 7.40Improved
use explicit 3D planar primitives instead of PlaneRecTR 2D queriesPlaneRecTR (ResNet50) · ScanNet valVI: 0.698 -> 0.610Improved
use explicit 3D planar primitives instead of PlaneRecTR 2D queriesPlaneRecTR (ResNet50) · ScanNet valplane parameter errors: 10.23 -> 7.77Improved
use PlaneRAS with pretrained depth priors instead of PlaneRecTR (Swin-B) for depthPlaneRecTR (Swin-B) · NYUv2 planar-region depthRel: 0.157 -> 0.165Hurt

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