Research map / Detection and segmentation / CNN features
Boundary-Aware 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 |
|---|---|---|---|
| represent object masks with a truncated distance map and decode masks beyond box boundaries (ablation) | BAIS instance segmentation · PASCAL VOC 2012 val | mAP (0.7): 44.58 -> 48.30 | Helped |
| represent object masks with a truncated distance map and decode masks beyond box boundaries (ablation) | BAIS instance segmentation · PASCAL VOC 2012 val | mAP (0.5): 64.97 -> 65.69 | Helped |
| train the cascade using 5 stages rather than 3 (ablation) | BAIS · PASCAL VOC 2012 val | mAP (0.7): 47.13 -> 48.30 | Helped |
| train the cascade using 5 stages rather than 3 (ablation) | BAIS · PASCAL VOC 2012 val | mAP (0.5): 65.51 -> 65.69 | Helped |
| use boundary-aware OMN masks instead of box-limited masks (ablation) | BAIS · Cityscapes val | mAP, IoU 0.7: 18.64 -> 19.10 | Helped |
| use boundary-aware OMN masks instead of box-limited masks (ablation) | BAIS · Cityscapes val | mAP, IoU 0.5: 31.05 -> 32.40 | Helped |
| use OMN segment proposals rather than MNC+score proposals | segment proposal generation · PASCAL VOC 2012 val, top 10 proposals | AR@10: 45.7 -> 47.8 | Helped |
| use OMN segment proposals rather than MNC+score proposals | segment proposal generation · PASCAL VOC 2012 val, top 100 proposals | AR@100: 49.1 -> 51.8 | Helped |
| use OMN proposals rather than Sharp-Mask proposals | segment proposal generation · PASCAL VOC 2012 val, top 1000 proposals | AR@1000: 56.5 vs 54.7 | Hurt |
Closest papers by meaning
- Instance-Aware Semantic Segmentation via Multi-Task Network Cascades (CVPR 2016 · similarity 0.91)
- Instance Segmentation With Mask-Supervised Polygonal Boundary Transformers (CVPR 2022 · similarity 0.90)
- SIM: Semantic-Aware Instance Mask Generation for Box-Supervised Instance Segmentation (CVPR 2023 · similarity 0.90)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (ICLR 2024 · similarity 0.89)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (CVPR 2024 · similarity 0.89)
- P2Seg: Pointly-supervised Segmentation via Mutual Distillation (ICLR 2024 · similarity 0.88)
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.