Research map / 3D vision / 3D generation
SceneCAD: Predicting Object Alignments and Layouts in RGB-D Scans
Armen Avetisyan, Tatiana Khanova, Christopher Choy, Denver Dash, Angela Dai, Matthias Nießner
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add object-object and object-layout relation inference (full) to sparse backbone (ablation) | Ours (sparse) · ScanNet Scan2CAD CAD alignment | alignment accuracy (avg.): 55.77 -> 61.24 | Improved |
| add object-object and object-layout relation inference (full) to sparse backbone (ablation) | Ours (sparse) · SUNCG CAD alignment | alignment accuracy (avg.): 52.31 -> 58.41 | Improved |
| add both relations to dense backbone (ablation) | Ours (dense) · ScanNet Scan2CAD CAD alignment | alignment accuracy (avg.): 51.05 -> 54.33 | Improved |
| add both relations to dense backbone (ablation) | Ours (dense) · SUNCG CAD alignment | alignment accuracy (avg.): 44.48 -> 50.41 | Improved |
| add object-object relations only to sparse backbone (ablation) | Ours (sparse) · ScanNet Scan2CAD CAD alignment | alignment accuracy (avg.): 55.77 -> 56.70 | Improved |
| add layout relations only to sparse backbone (ablation) | Ours (sparse) · ScanNet Scan2CAD CAD alignment | alignment accuracy (avg.): 55.77 -> 56.29 | Improved |
| add object-object relations only to sparse backbone (ablation) | Ours (sparse) · SUNCG CAD alignment | alignment accuracy (avg.): 52.31 -> 55.27 | Improved |
| add layout relations only to sparse backbone (ablation) | Ours (sparse) · SUNCG CAD alignment | alignment accuracy (avg.): 52.31 -> 57.12 | Improved |
| use sparse instead of dense convolutional backbone (ablation) | Ours full · ScanNet Scan2CAD CAD alignment | alignment accuracy (avg.): 54.33 -> 61.24 | Improved |
| use sparse instead of dense convolutional backbone (ablation) | Ours full · SUNCG CAD alignment | alignment accuracy (avg.): 50.41 -> 58.41 | Improved |
| add object-object relations to dense backbone (ablation) | Ours (dense) · ScanNet Scan2CAD CAD alignment | alignment accuracy (avg.): 51.05 -> 52.97 | Improved |
| add layout relations to dense backbone (ablation) | Ours (dense) · SUNCG CAD alignment | alignment accuracy (avg.): 44.48 -> 48.33 | Improved |
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.
- 3DMatch — 3DMatch: Learning Local Geometric Descriptors From RGB-D Reconstructions (CVPR 2017) · better on 1 of 1 shared result
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
- Scan2CAD: Learning CAD Model Alignment in RGB-D Scans (CVPR 2019 · similarity 0.91)
- End-to-End CAD Model Retrieval and 9DoF Alignment in 3D Scans (ICCV 2019 · similarity 0.90)
- FastCAD: Real-Time CAD Retrieval and Alignment from Scans and Videos (ECCV 2024 · similarity 0.89)
- Learning 3D Scene Priors With 2D Supervision (CVPR 2023 · similarity 0.88)
- Vid2CAD: CAD Model Alignment Using Multi-View Constraints From Videos (TPAMI 2022 · similarity 0.88)
- Learning Local RGB-to-CAD Correspondences for Object Pose Estimation (ICCV 2019 · similarity 0.87)
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.
