Research map / Detection and segmentation / CNN features
PolarMask: Single Shot Instance Segmentation With Polar Representation
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 |
|---|---|---|---|
| Increase the number of rays from 18 to 36 (ablation) | PolarMask · COCO minival | AP: 26.2 -> 27.7 | Helped |
| Increase the number of rays from 36 to 72 (ablation) | PolarMask · COCO minival | AP: 27.7 -> 27.6 | No clear effect |
| Replace Smooth-L1 loss with Polar IoU Loss (ablation) | PolarMask · COCO minival | AP: 25.1 -> 27.7 | Helped |
| Replace original centerness with Polar Centerness (ablation) | PolarMask · COCO minival | AP: 27.7 -> 29.1 | Helped |
| Add a box prediction branch (ablation) | PolarMask · COCO minival | AP: 27.5 -> 27.7 | No clear effect |
| Use ResNet-101 instead of ResNet-50 (ablation) | PolarMask · COCO minival | AP: 29.1 -> 30.4 | Helped |
| Use ResNeXt-101 instead of ResNet-50 (ablation) | PolarMask · COCO minival | AP: 29.1 -> 32.6 | Helped |
| Increase image scale from 400 to 800 (ablation) | PolarMask with ResNet-50-FPN · COCO minival | AP: 22.9 -> 29.1 | Helped |
| Increase image scale from 400 to 800 (ablation) | PolarMask with ResNet-50-FPN · COCO minival | FPS: 26.3 -> 17.2 | Hurt |
| Add DCN to the backbone and head (ablation) | PolarMask · COCO minival | AP: 29.1 -> 32.0 | Helped |
| Add multi-scale training (ablation) | PolarMask with ResNet-50 · COCO minival | AP: 29.1 -> 30.5 | Helped |
| Add Polar IoU Loss and Polar Centerness (ablation) | PolarMask · COCO 2017 val | AP: 25.1 -> 29.1 | Helped |
Closest papers by meaning
- CenterMask: Single Shot Instance Segmentation With Point Representation (CVPR 2020 · similarity 0.91)
- PolarNeXt: Rethink Instance Segmentation with Polar Representation (CVPR 2025 · similarity 0.90)
- BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation (CVPR 2020 · similarity 0.89)
- SOLOv2: Dynamic and Fast Instance Segmentation (NeurIPS 2020 · similarity 0.89)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (ICLR 2024 · similarity 0.87)
- MaskDiff: Modeling Mask Distribution with Diffusion Probabilistic Model for Few-Shot Instance Segmentation (AAAI 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.