AI Paper Atlas

Research map / 3D vision / Depth and stereo

Floorplan-Jigsaw: Jointly Estimating Scene Layout and Aligning Partial Scans

Cheng Lin, Changjian Li, Wenping Wang

ICCV 2019 · 3D vision › Depth and stereo · approach: Feed-forward 3D reconstruction · 20 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 closure term to the global placement energy (ablation)Floorplan-Jigsaw global placement · Global layout placement (101 scenes)ACC_global (%): 22.8 -> 85.1Improved
add the layout complexity term to the global placement energy (ablation)Floorplan-Jigsaw global placement · Global layout placement (101 scenes)ACC_global (%): 67.5 -> 85.1Improved
add the boundary similarity term to the global placement energy (ablation)Floorplan-Jigsaw global placement · Global layout placement (101 scenes)ACC_global (%): 80.2 -> 85.1Improved
use layout-based alignment instead of feature-descriptor registration (3DMatch) for small-overlap scanspartial scan alignment · SUNCG pose estimationrotation error (deg): 43.41 -> 8.79Improved
use layout-based alignment instead of FPFH for small-overlap scanspartial scan alignment · SUNCG pose estimationtranslation error (%): 29.12 -> 9.15Improved
use layout-based alignment instead of Manhattan-world OPVO for small-overlap scanspartial scan alignment · SUNCG pose estimationtranslation error (%): 20.04 -> 9.15Improved
use the floorplan path method instead of RAPTER for structured modelinglayout reconstruction · Layout reconstruction (complete point clouds)max error (%): 7.84 -> 2.57Improved
use the floorplan path method instead of MW Modeler for structured modelinglayout reconstruction · Layout reconstruction (complete point clouds)average error (%): 1.22 -> 0.90Improved
increase the complexity penalty lambda (ablation)local layout estimation · Partial layout estimation (401 fragments)layout detailMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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