Research map / 3D vision / 3D generation
Learning 3D Object Shape and Layout Without 3D Supervision
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Train using 5 rather than 2 views (ablation) | USL · Scene-Shapes val | F1@0.1m: 62.9 -> 70.9 | Helped |
| Replace RoIAlign with RoIMap (ablation) | USL · Scene-Shapes val | Mask 2D IoU (views): 40.9 -> 46.7 | Helped |
| Add the distance transform loss (ablation) | USL · Scene-Shapes val | Mask 2D IoU (views): 37.1 -> 46.7 | Helped |
| Train using 5 rather than 2 views (ablation) | USL · Hypersim val | Mask 2D IoU (views): 0.33 -> 0.34 | Helped |
| Replace RoIAlign with RoIMap (ablation) | USL · Hypersim val | Mask 2D IoU (views): 0.20 -> 0.34 | Helped |
| Add the distance transform loss (ablation) | USL · Hypersim val | Mask 2D IoU (views): 0.28 -> 0.34 | Helped |
| Predict full shapes rather than fixed spheres (ablation) | USL with oracle depth · Hypersim val | Mask 2D IoU (views): 0.45 -> 0.53 | Helped |
Closest papers by meaning
- Learning 3D Scene Priors With 2D Supervision (CVPR 2023 · similarity 0.93)
- Factoring Shape, Pose, and Layout From the 2D Image of a 3D Scene (CVPR 2018 · similarity 0.89)
- Front2Back: Single View 3D Shape Reconstruction via Front to Back Prediction (CVPR 2020 · similarity 0.89)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.88)
- Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space (CVPR 2025 · similarity 0.88)
- Disentangled 3D Scene Generation with Layout Learning (ICML 2024 · similarity 0.88)
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.