Research map / Detection and segmentation / Weakly supervised detection and segmentation
Learning Instance Activation Maps for Weakly Supervised 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 |
|---|---|---|---|
| Replace PRM proposal retrieval with IAM extent filling | Weakly supervised instance segmentation · PASCAL VOC 2012 | mAP r 0.25 (%): 44.3 -> 45.9 | Helped |
| Replace PRM proposal retrieval with IAM extent filling | Weakly supervised instance segmentation · PASCAL VOC 2012 | mAP r 0.5 (%): 26.8 -> 28.8 | Helped |
| Replace PRM proposal retrieval with IAM extent filling | Weakly supervised instance segmentation · PASCAL VOC 2012 | mAP r 0.75 (%): 9.0 -> 11.9 | Helped |
| Replace PRM proposal retrieval with IAM extent filling | Weakly supervised instance segmentation · PASCAL VOC 2012 | ABO: 37.6 -> 41.9 | Helped |
| Use random sampling number 1 instead of 5 (ablation) | IAM · PASCAL VOC 2012 | mAP r 0.75 (%): 11.9 -> 10.4 | Hurt |
| Use random sampling number 9 instead of 5 (ablation) | IAM · PASCAL VOC 2012 | mAP r 0.5 (%): 28.8 -> 27.8 | Hurt |
| Use random sampling number 9 instead of 5 (ablation) | IAM · PASCAL VOC 2012 | mAP r 0.25 (%): 45.9 -> 45.7 | Hurt |
| Use random sampling number 9 instead of 5 (ablation) | IAM · PASCAL VOC 2012 | mAP r 0.75 (%): 11.9 -> 10.5 | Hurt |
| Use random sampling number 9 instead of 5 (ablation) | IAM · PASCAL VOC 2012 | ABO: 41.9 -> 41.7 | Hurt |
| Apply VOC12-trained IAM extent filling without target-dataset fine-tuning | Weakly supervised object localization · CUB-200-2011 test | Localization error (%): 53.36 -> 52.21 | Helped |
| Apply VOC12-trained IAM to salient object detection without fine-tuning | Salient object detection · THUR | Mean F-measure (Fβ): 0.7322 -> 0.7364 | Helped |
| Apply VOC12-trained IAM to salient object detection without fine-tuning | Salient object detection · MSRA-B | Mean F-measure (Fβ): 0.8973 -> 0.8643 | Hurt |
Closest papers by meaning
- Weakly Supervised Instance Segmentation Using Class Peak Response (CVPR 2018 · similarity 0.92)
- Sparse Instance Activation for Real-Time Instance Segmentation (CVPR 2022 · similarity 0.90)
- Weakly Supervised Learning of Instance Segmentation With Inter-Pixel Relations (CVPR 2019 · similarity 0.90)
- UniT: Unified Knowledge Transfer for Any-Shot Object Detection and Segmentation (CVPR 2021 · similarity 0.90)
- WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels (CVPR 2025 · similarity 0.88)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (ICLR 2024 · 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.