Research map / Detection and segmentation / CNN features
Learning to Detect Every Thing in an Open World
Kuniaki Saito, Ping Hu, Trevor Darrell, Kate Saenko
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 BackErase synthetic-background augmentation with decoupled training to Mask R-CNN | Mask R-CNN · VOC->COCO non-VOC | AR100: 10.9 -> 30.8 | Improved |
| add mask loss on real images to detection loss on synthetic images (decoupled training) versus synthetic-only training (ablation) | LDET · VOC->COCO non-VOC | AR100: 4.3 -> 30.8 | Improved |
| add mask loss on synthetic data to synthetic-only detection training (ablation) | Synth Only · VOC->COCO non-VOC | AR100: 4.3 -> 9.5 | Improved |
| add synthetic-background augmentation on top of plain real-image training (ablation) | Plain Mask R-CNN · VOC->COCO non-VOC | AR10: 8.8 -> 18.2 | Improved |
| reduce background crop region from 1/2 to 1/8 of image side (ablation) | LDET · VOC->COCO non-VOC | AR100: 26.2 -> 30.8 | Improved |
| reduce background crop region from 1/8 to 1/16 of image side (ablation) | LDET · VOC->COCO non-VOC | AR100: 30.8 -> 31.8 | Improved |
| reduce background crop region from 1/8 to 1/16 of image side (ablation) | LDET · VOC->COCO non-VOC | AR10: 18.2 -> 17.5 | Hurt |
| use DTD texture patches as external background instead of cropped self-image (ablation) | LDET · VOC->COCO non-VOC | box AR: 26.7 (LDET-3.1) | Hurt |
| add LDET to RetinaNet | RetinaNet · VOC->COCO non-VOC | AR100: 17.8 -> 31.0 | Improved |
| add LDET to TensorMask | TensorMask · VOC->COCO non-VOC | AR100: 19.7 -> 31.1 | Improved |
| add LDET to Mask R-CNN trained on full COCO | Mask R-CNN · COCO->UVO | box AR: 42.3 -> 47.5 | Improved |
| add LDET to Mask R-CNN trained on VOC classes | Mask R-CNN · COCO->UVO | box AR: 30.0 -> 43.7 | Improved |
Closest papers by meaning
- SegPrompt: Boosting Open-World Segmentation via Category-Level Prompt Learning (ICCV 2023 · similarity 0.90)
- Detecting Everything in the Open World: Towards Universal Object Detection (CVPR 2023 · similarity 0.89)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (ICLR 2024 · similarity 0.88)
- Detecting Open World Objects via Partial Attribute Assignment (CVPR 2025 · similarity 0.87)
- v-CLR: View-Consistent Learning for Open-World Instance Segmentation (CVPR 2025 · similarity 0.87)
- Exploring Orthogonality in Open World Object Detection (CVPR 2024 · similarity 0.86)
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.
