Research map / Label-efficient and robust learning / Domain adaptation and generalization
SRDA: Generating Instance Segmentation Annotation via Scanning, Reasoning and Domain Adaptation
Wenqiang Xu, Yonglu Li, Cewu Lu
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 geometry-guided loss (ablation) | GeoGAN + Mask R-CNN · Instance-60K shelf | mAP@0.5: 48.52 -> 66.31 | Improved |
| add structure (PMSE) loss (ablation) | GeoGAN + Mask R-CNN · Instance-60K shelf | mAP@0.5: 27.33 -> 66.31 | Improved |
| add the physics reasoning system for layouts (ablation) | SRDA pipeline · Instance-60K shelf | mAP@0.5: 15.21 -> 66.31 | Improved |
| add geometry-guided loss (ablation) | GeoGAN + Mask R-CNN · Instance-60K desk | mAP@0.5: 63.99 -> 82.07 | Improved |
| add structure (PMSE) loss (ablation) | GeoGAN + Mask R-CNN · Instance-60K desk | mAP@0.5: 45.05 -> 82.07 | Improved |
| add the physics reasoning system for layouts (ablation) | SRDA pipeline · Instance-60K desk | mAP@0.5: 18.36 -> 82.07 | Improved |
| add geometry-guided loss (ablation) | GeoGAN + Mask R-CNN · Instance-60K tote | mAP@0.5: 64.22 -> 82.69 | Improved |
| add structure (PMSE) loss (ablation) | GeoGAN + Mask R-CNN · Instance-60K tote | mAP@0.5: 46.44 -> 82.69 | Improved |
| add the physics reasoning system for layouts (ablation) | SRDA pipeline · Instance-60K tote | mAP@0.5: 20.05 -> 82.69 | Improved |
| add geometry path and geometry loss (ablation) | GeoGAN · Instance-60K shelf | mAP@0.5: 43.68 -> 66.31 | Improved |
| add normal map input to geometry path (full: depth, normal, seg) (ablation) | GeoGAN · Instance-60K shelf | mAP@0.5: 62.87 -> 66.31 | Improved |
| add segmentation input to geometry path (ablation) | GeoGAN · Instance-60K shelf | mAP@0.5: 58.05 -> 66.31 | Improved |
Closest papers by meaning
- 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans (CVPR 2019 · similarity 0.86)
- SAI3D: Segment Any Instance in 3D Scenes (CVPR 2024 · similarity 0.86)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.85)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 2025 · similarity 0.85)
- A Generative Approach at the Instance-Level for Image Segmentation Under Limited Training Data Conditions (Student Abstract) (AAAI 2025 · similarity 0.85)
- UnScene3D: Unsupervised 3D Instance Segmentation for Indoor Scenes (CVPR 2024 · similarity 0.84)
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.
