Research map / Detection and segmentation / Open-set and few-shot detection
One-Shot Segmentation in Clutter
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 |
|---|---|---|---|
| Add proposal segmentation that masks background before selecting a match | Siamese U-Net baseline · cluttered Omniglot, 32–256 characters | IoU (%): 62.4 -> 65.6; 48.1 -> 52.8; 39.3 -> 44.8; 38.4 -> 43.7 | Helped |
| Add proposal segmentation that masks background before selecting a match | Siamese U-Net baseline · cluttered Omniglot, 4–16 characters | IoU (%): 97.1 -> 95.8; 92.1 -> 90.5; 79.8 -> 79.3 | Hurt |
| Add segmentation before discriminating candidate locations | Cluttered discriminator · cluttered Omniglot, 4–256 characters | localization accuracy (%): 97.0 -> 97.4; 92.1 -> 94.1; 82.2 -> 87.0; 67.1 -> 77.5; 54.7 -> 66.1; 44.2 -> 58.5; 41.3 -> 57.7 | Helped |
| Add the target embedding to the proposal decoder input (ablation) | MaskNet proposal network · cluttered Omniglot, 32–256 characters | IoU (%): 52.7 -> 65.6; 39.0 -> 52.8; 30.7 -> 44.8; 27.3 -> 43.7 | Helped |
| Use ground-truth segmentation masks to pre-segment candidate crops | Cluttered discriminator · cluttered Omniglot, 4–256 characters | localization accuracy (%): 97.0 -> 99.6; 92.1 -> 99.2; 82.2 -> 98.9; 67.1 -> 98.2; 54.7 -> 97.8; 44.2 -> 96.9; 41.3 -> 96.2 | Helped |
| Increase the number of distractors | Siamese U-Net · cluttered Omniglot, 4 to 256 characters | IoU (%): 97.1 -> 38.4 | Hurt |
| Replace learned discrimination with template matching | Pre-segmented discriminator · cluttered Omniglot, 4–256 characters | IoU (%): 62.2, 50.4, 41.7, 36.9, 32.6, 29.0, 28.6 | Hurt |
Closest papers by meaning
- What Ails One-Shot Image Segmentation: A Data Perspective (NeurIPS 2021 · similarity 0.88)
- One-Shot Video Object Segmentation (CVPR 2017 · similarity 0.87)
- MaskDiff: Modeling Mask Distribution with Diffusion Probabilistic Model for Few-Shot Instance Segmentation (AAAI 2024 · similarity 0.83)
- SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentation (NeurIPS 2025 · similarity 0.83)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (ICLR 2024 · similarity 0.83)
- A Generative Approach at the Instance-Level for Image Segmentation Under Limited Training Data Conditions (Student Abstract) (AAAI 2025 · similarity 0.82)
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.