AI Paper Atlas

Research map / 3D vision / 3D generation

Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries

Yuanwen Yue, Theodora Kontogianni, Konrad Schindler, Francis Engelmann

CVPR 2023 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 48 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
Predict rooms as a set of ordered vertex sequences with two-level queries rather than single-level queries (ablation)RoomFormer · Structured3D valRoom precision: 74.4 -> 96.5Improved
Use multi-scale feature maps instead of only the last-stage feature map (ablation)RoomFormer · Structured3D valRoom precision: 93.9 -> 96.5Improved
Iteratively refine polygon coordinates and use updated positions as reference points (ablation)RoomFormer · Structured3D valRoom precision: 94.8 -> 96.5Improved
Add coordinate regression loss alongside classification and rasterization losses (ablation)RoomFormer · Structured3D valRoom precision: 84.3 -> 96.5Improved
Add polygon rasterization loss (ablation)RoomFormer · Structured3D valRoom precision: 96.0 -> 96.5Improved
Allow self-attention between vertices across different polygons (ablation)RoomFormer · Structured3D valRoom F1: 94.8 -> 95.9Improved
Increase room-level and corner-level query counts from M=15, N=30 to M=20, N=40 (ablation)RoomFormer · Structured3D valinference time: 8.4 -> 10.8Hurt
Use separate room and line decoders with two-level line queries (ablation)RoomFormer semantic-rich extension · Structured3D test, semantic-richRoom type F1: 70.7 -> 71.4Improved
Use single-level rather than two-level queries in the separate line decoder (ablation)RoomFormer semantic-rich extension · Structured3D test, semantic-richRoom type F1: 71.4 -> 74.4Improved
Train using tight room layouts rather than the original room layout representation (ablation)RoomFormer · Structured3D testAngle F1: 81.2 -> 84.2Improved
Train using tight room layouts rather than the original room layout representation (ablation)RoomFormer · Structured3D testRoom F1: 97.3 -> 96.7Hurt
Replace the original RoomFormer with holistic polygon prediction instead of a corner-and-edge pipelineRoomFormer versus HEAT · SceneCAD val, trained on Structured3DRoom IoU: 52.5 -> 74.0Improved

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