Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Learning of Instance Segmentation With Inter-Pixel Relations
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 |
|---|---|---|---|
| add class-boundary propagation to CAMs (ablation) | CAM-based pseudo instance labels · PASCAL VOC 2012 train | AP^r_50: 8.6 -> 34.1 | Helped |
| add a displacement field to CAMs with class-boundary propagation (ablation) | CAM-based pseudo instance labels · PASCAL VOC 2012 train | AP^r_50: 34.1 -> 37.7 | Helped |
| propagate CAM attention using predicted semantic affinities (ablation) | CAM-based pseudo semantic labels · PASCAL VOC 2012 train | mIoU: 48.3 -> 66.5 | Helped |
| use IRNet rather than AffinityNet for semantic propagation | CAM-based pseudo semantic labels · PASCAL VOC 2012 train | mIoU: 59.3 -> 66.5 | Helped |
| train with image-level labels and IRNet-generated pseudo instance labels | Ours-ResNet50 · PASCAL VOC 2012 val | AP^r_50: 26.8 -> 46.7 | Helped |
| train with image-level labels and IRNet-generated pseudo instance labels | Ours-ResNet50 · PASCAL VOC 2012 val | AP^r_50: 44.8 -> 46.7 | Helped |
| train DeepLab with IRNet-generated pseudo semantic labels | Ours-ResNet50 · PASCAL VOC 2012 val | mIoU: 61.4 -> 63.5 | Helped |
Closest papers by meaning
- Weakly Supervised Instance Segmentation Using Class Peak Response (CVPR 2018 · similarity 0.90)
- Learning Instance Activation Maps for Weakly Supervised Instance Segmentation (CVPR 2019 · similarity 0.90)
- WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels (CVPR 2025 · similarity 0.89)
- Extreme Point Supervised Instance Segmentation (CVPR 2024 · similarity 0.87)
- P2Seg: Pointly-supervised Segmentation via Mutual Distillation (ICLR 2024 · similarity 0.87)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 2025 · similarity 0.87)
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.