AI Paper Atlas

Research map / Point clouds / Segmentation and detection

BuildingNet: Learning To Label 3D Buildings

Pratheba Selvaraju, Mohamed Nabail, Marios Loizou, Maria Maslioukova, Melinos Averkiou, Andreas Andreou, Siddhartha Chaudhuri, Evangelos Kalogerakis

ICCV 2021 · Point clouds › Segmentation and detection · approach: Scene completion and instance segmentation · 47 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 the building GNN over subgroup graph on MinkowskiNet featuresMinkNet2Sub · BuildingNet-Mesh (no color)Part IoU (%): 33.1 -> 40.0Improved
add the building GNN over subgroup graph on MinkowskiNet featuresMinkNet2Sub · BuildingNet-Mesh (with color)Part IoU (%): 37.0 -> 42.6Improved
add the building GNN over subgroup graph on PointNet++ featuresPointNet2Sub · BuildingNet-Mesh (with color)Part IoU (%): 16.1 -> 31.5Improved
replace graph cuts with the GNNMinkNet-GC · BuildingNet-Mesh (no color)Part IoU (%): 29.9 -> 40.0Improved
aggregate point predictions per mesh subgroup instead of per triangleMinkNet · BuildingNet-Mesh (no color)Part IoU (%): 28.8 -> 33.1Improved
add RGB color inputMinkNet · BuildingNet-PointsPart IoU (%): 26.9 -> 29.9Improved
use average pooling instead of max pooling over subgroup points (ablation)MinkNet2Sub · BuildingNet-Mesh (no color)Part IoU (%): 30.4 -> 33.1Improved
use average pooling instead of max pooling over triangle points (ablation)MinkNet2Triangle · BuildingNet-Mesh (with color)Part IoU (%): 31.5 -> 32.8Improved
use weighted cross-entropy instead of unweighted cross-entropy (ablation)MinkNet · BuildingNet-Points (no color)Part IoU (%): 24.5 -> 26.9Improved
use weighted cross-entropy instead of alpha-balanced focal loss (ablation)MinkNet · BuildingNet-Points (no color)Part IoU (%): 22.3 -> 26.9Improved
use class-balanced loss instead of weighted cross-entropy (ablation)MinkNet · BuildingNet-Points (with color)Part IoU (%): 29.9 -> 30.4No clear effect
add each edge type (proximity, support, similarity, containment) to node features (ablation)BuildingGNN · BuildingNet-MeshPart IoU (%)Improved

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