Research map / Detection and segmentation / CNN features
Iterative Instance Segmentation
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 |
|---|---|---|---|
| Use iterative prediction and correction instead of single-step Hypercolumn prediction | Instance segmentation · PASCAL VOC 2012 val, raw pixel-wise, 50% overlap | mAP^r: 56.1 -> 60.1 | Helped |
| Use iterative prediction and correction instead of single-step Hypercolumn prediction | Instance segmentation · PASCAL VOC 2012 val, raw pixel-wise, 70% overlap | mAP^r: 29.4 -> 38.7 | Helped |
| Add superpixel projection to raw pixel-wise predictions (ablation) | Proposed Method · PASCAL VOC 2012 val, 50% overlap | mAP^r: 60.1 -> 60.3 | Helped |
| Add superpixel projection to raw pixel-wise predictions (ablation) | Proposed Method · PASCAL VOC 2012 val, 70% overlap | mAP^r: 38.7 -> 40.2 | Helped |
| Add rescoring after superpixel projection (ablation) | Proposed Method · PASCAL VOC 2012 val, 50% overlap | mAP^r: 60.3 -> 63.6 | Helped |
| Add rescoring after superpixel projection (ablation) | Proposed Method · PASCAL VOC 2012 val, 70% overlap | mAP^r: 40.2 -> 43.3 | Helped |
| Apply superpixel projection to raw pixel-wise predictions (ablation) | Proposed Method · PASCAL VOC 2012 val, across overlap thresholds and categories | mAP^r: 60.1 -> 60.3; 38.7 -> 40.2 | Mixed |
Closest papers by meaning
- Pixelwise Instance Segmentation With a Dynamically Instantiated Network (CVPR 2017 · similarity 0.89)
- WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels (CVPR 2025 · similarity 0.88)
- InstanceCut: From Edges to Instances With MultiCut (CVPR 2017 · similarity 0.88)
- P2Seg: Pointly-supervised Segmentation via Mutual Distillation (ICLR 2024 · similarity 0.87)
- Minimizing Labeled, Maximizing Unlabeled: An Image-Driven Approach for Video Instance Segmentation (CVPR 2025 · similarity 0.86)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 2025 · 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.