AI Paper Atlas

Research map / Point clouds / Shape and surface completion

Edges Compete for Trust: Group Relative Edge Optimization for Building Reconstruction from Point Clouds

Yujun Liu, Ruisheng Wang, Xiang Ao, Haoyuan Shen, Kuihao Wang, Kun Zhou, Qingquan Li

CVPR 2026 · Point clouds › Shape and surface completion · approach: Unsupervised shape correspondence

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 GREO training strategy to EdgeDiffEdgeDiff · Building3D TallinnWED: 0.255 -> 0.225Improved
add GREO training strategy to PBWRPBWR · Building3D TallinnWED: 0.271 -> 0.250Improved
add GREO to EdgeDiff on Entry-levelEdgeDiff · Building3D Entry-levelWED: 0.115 -> 0.108Improved
add GREO to PBWR on Entry-levelPBWR · Building3D Entry-levelWED: 0.145 -> 0.130Improved
add reward-based cross-entropy loss L_CE (ablation)EdgeDiff · Building3D TallinnWED: 0.255 -> 0.242Improved
add entropy regularization on top of L_CE (ablation)EdgeDiff · Building3D TallinnWED: 0.242 -> 0.225Improved
add entropy regularization on top of L_CE (ablation)PBWR · Building3D TallinnWED: 0.262 -> 0.250Improved
use z-score instead of no normalization for rewards (ablation)EdgeDiff · Building3D TallinnWED: 0.249 -> 0.225Improved
use z-score instead of min-max normalization (ablation)EdgeDiff · Building3D TallinnWED: 0.236 -> 0.225Improved
use absolute reward without normalization (ablation)PBWR · Building3D TallinnWED: 0.271 -> 0.266No clear effect
combine Hausdorff, cosine and center terms in reward instead of Hausdorff only (ablation)EdgeDiff · Building3D TallinnWED: 0.247 -> 0.225Improved
add center term to Hausdorff+cosine reward (ablation)EdgeDiff · Building3D TallinnWED: 0.237 -> 0.225Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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