Research map / Detection and segmentation / Weakly supervised detection and segmentation
Self-Erasing Network for Integral Object Attention
Read the paperCheck 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 SeeNet instead of DCSP with ResNet101 | Weakly-supervised semantic segmentation · PASCAL VOC 2012 val | mIoU, w/ CRF: 60.8% -> 63.1% | Helped |
| Use SeeNet instead of DCSP with VGG16 | Weakly-supervised semantic segmentation · PASCAL VOC 2012 val | mIoU, w/ CRF: 58.6% -> 61.1% | Helped |
| Add self-erasing strategies to the ACoL-style setting (ablation) | SeeNet attention network · PASCAL VOC 2012 val | mIoU, w/o CRF: 56.1% -> 57.3% | Helped |
| Add conditional sign-flipping in C-ReLU instead of zeroing background-associated feature units (ablation) | SeeNet · PASCAL VOC 2012 val | mIoU, w/o CRF: 55.8% -> 57.3% | Helped |
| Add the background-prediction branch S_C (ablation) | SeeNet · PASCAL VOC 2012 | attention-map quality | Helped |
Closest papers by meaning
- Object Region Mining With Adversarial Erasing: A Simple Classification to Semantic Segmentation Approach (CVPR 2017 · similarity 0.87)
- Erasing Integrated Learning: A Simple Yet Effective Approach for Weakly Supervised Object Localization (CVPR 2020 · similarity 0.86)
- Tell Me Where to Look: Guided Attention Inference Network (CVPR 2018 · similarity 0.84)
- Scaling Local Self-Attention for Parameter Efficient Visual Backbones (CVPR 2021 · similarity 0.84)
- Attention is not all you need: pure attention loses rank doubly exponentially with depth (ICML 2021 · similarity 0.84)
- Attentive Eraser: Unleashing Diffusion Model’s Object Removal Potential via Self-Attention Redirection Guidance (AAAI 2025 · similarity 0.83)
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.