Research map / Detection and segmentation / Open-set and few-shot detection
Unsupervised Object Localization: Observing the Background To Discover Objects
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 background discovery and train a lightweight foreground head rather than the strongest listed baseline | FOUND · VOC07 trainval | CorLoc: 72.3 -> 72.5 | Helped |
| Use background discovery and train a lightweight foreground head rather than the strongest listed baseline | FOUND · VOC12 trainval | CorLoc: 75.3 -> 76.1 | Helped |
| Use background discovery and train a lightweight foreground head rather than the strongest listed baseline | FOUND · COCO20k | CorLoc: 67.2 -> 62.9 | Hurt |
| Use FOUND multi rather than SelfMask with bilateral solver | FOUND · DUTS-TE | IoU: .660 -> .663 | Hurt |
| Use FOUND multi rather than SelfMask with bilateral solver | FOUND · ECSSD | IoU: .818 -> .813 | Hurt |
| Use FOUND for single-mask retrieval rather than SelfMask | FOUND · PASCAL VOC12 7cls, single mask | mIoU: 56.6 -> 56.1 | Hurt |
| Add attention-head reweighting (ablation) | FOUND background discovery · ECSSD coarse masks, multi | IoU: .627 -> .637 | Helped |
| Add attention-head reweighting (ablation) | FOUND background discovery · ECSSD coarse masks, single | IoU: .671 -> .679 | Helped |
| Add bilateral solver as post-processing (ablation) | FOUND coarse masks, single · ECSSD | IoU: .671 -> .709 | Helped |
| Add bilateral solver during training (ablation) | FOUND · ECSSD trained masks, multi | IoU: .790 -> .807 | Helped |
| Also apply bilateral solver as post-processing (ablation) | FOUND trained masks, multi · ECSSD | IoU: .807 -> .813 | Helped |
| Train the segmentation head on FOUND coarse foreground masks (ablation) | FOUND · ECSSD trained masks, multi | IoU | Helped |
Closest papers by meaning
- Unsupervised Semantic Segmentation with Self-supervised Object-centric Representations (ICLR 2023 · similarity 0.93)
- Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks (AAAI 2019 · similarity 0.91)
- Self-Supervised Learning of Object Parts for Semantic Segmentation (CVPR 2022 · similarity 0.91)
- HEAP: Unsupervised Object Discovery and Localization with Contrastive Grouping (AAAI 2024 · similarity 0.88)
- unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning (ICML 2025 · similarity 0.88)
- Segment Anything without Supervision (NeurIPS 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.