AI Paper Atlas

Research map / Point clouds / Shape and surface completion

CAD-Recode: Reverse Engineering CAD Code from Point Clouds

Danila Rukhovich, Elona Dupont, Dimitrios Mallis, Kseniya Cherenkova, Anis Kacem, Djamila Aouada

ICCV 2025 · Point clouds › Shape and surface completion · approach: Unsupervised shape correspondence · 60 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
train on procedural 160k data instead of DeepCAD 160k (ablation)CAD-Recode · DeepCAD testmean Chamfer distance: 1.98 -> 0.54Improved
train on procedural 160k data instead of DeepCAD 160k (ablation)CAD-Recode · DeepCAD testIoU: 80.7 -> 88.3Improved
train on procedural 160k data instead of DeepCAD 160k (ablation)CAD-Recode · Fusion360 testIoU: 67.6 -> 82.0Improved
scale procedural training data from 160k to 1M (ablation)CAD-Recode · DeepCAD testIoU: 88.3 -> 92.0Improved
scale procedural training data from 160k to 1M (ablation)CAD-Recode · CC3D (real scans)IoU: 69.0 -> 74.2Improved
add test-time sampling over multiple point cloud samplings (ablation)CAD-Recode · DeepCAD testinvalidity ratio: 4.9 -> 0.4Improved
add test-time sampling (ablation)CAD-Recode · CC3D (real scans)invalidity ratio: 16.8 -> 0.3Improved
add test-time sampling (ablation)CAD-Recode · DeepCAD testmean Chamfer distance: 0.75 -> 0.30Improved
increase input points from 128 to 256 (0.5B) (ablation)CAD-Recode · DeepCAD testIoU: 89.9 -> 90.6Improved
increase LLM size from 0.5B to 1.5B (256 points) (ablation)CAD-Recode · DeepCAD testIoU: 90.6 -> 92.0Improved
increase LLM size from 0.5B to 1.5B (256 points) (ablation)CAD-Recode · CC3D (real scans)median Chamfer distance: 0.36 -> 0.31Improved
CAD-Recode vs CAD-SIGNet on real scansCAD-Recode · CC3D (real scans)median Chamfer distance: 2.90 -> 0.31Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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