Research map / 3D vision / Face and shape
Factoring Shape, Pose, and Layout From the 2D Image of a 3D Scene
Read the paperCheck 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 |
|---|---|---|---|
| add context features from the full image and bounding box coordinates (ablation) | Base object predictor · SUNCG, ground-truth boxes | Shape median IoU: 0.27 -> 0.31 | Helped |
| add context features from the full image and bounding box coordinates (ablation) | Base object predictor · SUNCG, ground-truth boxes | Rotation median error: 7.69 -> 5.44 | Helped |
| add context features from the full image and bounding box coordinates (ablation) | Base object predictor · SUNCG, ground-truth boxes | Translation median error: 0.47 -> 0.38 | Helped |
| add context features from the full image and bounding box coordinates (ablation) | Base object predictor · SUNCG, ground-truth boxes | Scale median error: 0.19 -> 0.15 | Helped |
| use rotation classification instead of regression (ablation) | Base object predictor · SUNCG, ground-truth boxes | Rotation error < 30 rate: 48.1% -> 75.2% | Helped |
| fine-tune the shape decoder (ablation) | Base object predictor · SUNCG, ground-truth boxes | Shape median IoU: 0.31 -> 0.41 | Helped |
| fine-tune the shape decoder (ablation) | Base object predictor · SUNCG, ground-truth boxes | Translation median error: 0.38 -> 0.42 | Hurt |
Closest papers by meaning
- Learning 3D Object Shape and Layout Without 3D Supervision (CVPR 2022 · similarity 0.89)
- DeLay: Robust Spatial Layout Estimation for Cluttered Indoor Scenes (CVPR 2016 · similarity 0.89)
- 3D Shape Attributes (CVPR 2016 · similarity 0.89)
- Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes From a Single Image (CVPR 2020 · similarity 0.88)
- Learning 3D Scene Priors With 2D Supervision (CVPR 2023 · similarity 0.88)
- From Programs to Poses: Factored Real-World Scene Generation via Learned Program Libraries (NeurIPS 2025 · 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.