Research map / Detection and segmentation / Open-set and few-shot detection
ScaleNet: Guiding Object Proposal Generation in Supermarkets and Beyond
Siyuan Qiao, Wei Shen, Weichao Qiu, Chenxi Liu, Alan Yuille
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 ScaleNet scale prediction to SharpMask-ft | SharpMask-ft · Real-Near supermarket, 100 proposals | average recall: 0.249 -> 0.396 | Improved |
| add ScaleNet scale prediction to SharpMask-ft | SharpMask-ft · Real-Far supermarket, 100 proposals | average recall: 0.224 -> 0.298 | Improved |
| add ScaleNet scale prediction to SharpMask-ft | SharpMask-ft · Real-Far supermarket, 1000 proposals | average recall: 0.526 -> 0.589 | Improved |
| add ScaleNet scale prediction to DeepMask-ft | DeepMask-ft · Real-Near supermarket, 100 proposals | average recall: 0.231 -> 0.373 | Improved |
| add ScaleNet scale prediction to DeepMask-ft | DeepMask-ft · Real-Far supermarket, 1000 proposals | average recall: 0.497 -> 0.561 | Improved |
| train on MS COCO plus the synthetic supermarket dataset (-ft) (ablation) | SharpMask · Real-Far supermarket, 100 proposals | average recall: 0.191 -> 0.224 | Improved |
| train on MS COCO plus the synthetic supermarket dataset (-ft) (ablation) | DeepMask · Real-Near supermarket, 1000 proposals | average recall: 0.488 -> 0.533 | Improved |
| train on MS COCO plus the synthetic supermarket dataset (-ft) (ablation) | SharpMask · Real-Near supermarket, 1000 proposals | average recall: 0.518 -> 0.567 | Improved |
| add ScaleNet to SharpMask on natural images | SharpMask · MS COCO val box proposals | AR@1k: 0.482 -> 0.557 | Improved |
| add ScaleNet to SharpMask-Ours on natural images | SharpMask-Ours · MS COCO val box proposals | AR@100: 0.392 -> 0.439 | Improved |
| add ScaleNet to SharpMask-Ours at 10 proposals | SharpMask-Ours · MS COCO val segmentation proposals | AR@10: 0.177 -> 0.177 | No clear effect |
| add ScaleNet to SharpMask-Ours (box recall at IoU 0.5, 1000 proposals) | SharpMask · MS COCO val box proposals | recall at IoU 0.5: 0.714 -> 0.843 | Improved |
Closest papers by meaning
- Toward Scale-Invariance and Position-Sensitive Region Proposal Networks (ECCV 2018 · similarity 0.86)
- MVTec D2S: Densely Segmented Supermarket Dataset (ECCV 2018 · similarity 0.84)
- ScaleNet: A Shallow Architecture for Scale Estimation (CVPR 2022 · similarity 0.84)
- Object-Proposal Evaluation Protocol is 'Gameable' (CVPR 2016 · similarity 0.84)
- GenScale: A Benchmark for Relative Object Scale in Image Generation and Editing (NeurIPS 2026 · similarity 0.83)
- Scaling Laws of Synthetic Images for Model Training ... for Now (CVPR 2024 · similarity 0.83)
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.
