Research map / Detection and segmentation / Open-set and few-shot detection
Few-Shot Object Detection via Association and DIscrimination
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 FADI instead of TFA | Few-shot object detection · PASCAL VOC Novel Split 1, 1-shot | novel AP50: 39.8 -> 50.3 | Helped |
| Use FADI instead of TFA | Few-shot object detection · MS COCO, 10-shot | nAP: 10.0 -> 12.2 | Helped |
| Use FADI rather than FSCE | Few-shot object detection · PASCAL VOC Novel Split 1, 5-shot | novel AP50: 61.9 -> 59.3 | Hurt |
| Add association (ablation) | TFA · PASCAL VOC Novel Split 1, 1-shot | nAP50: 41.3 -> 42.4 | Helped |
| Add disentangling (ablation) | TFA · PASCAL VOC Novel Split 1, 1-shot | nAP50: 41.3 -> 42.2 | Helped |
| Add association and disentangling together (ablation) | TFA · PASCAL VOC Novel Split 1, 1-shot | nAP50: 41.3 -> 44.9 | Helped |
| Add set-specialized margin loss (ablation) | TFA · PASCAL VOC Novel Split 1, 1-shot | nAP50: 41.3 -> 46.3 | Helped |
| Add set-specialized margin loss (ablation) | Association + disentangling · PASCAL VOC Novel Split 1, 1-shot | nAP50: 44.9 -> 50.3 | Helped |
| Apply standard CosFace or ArcFace margin loss across samples (ablation) | TFA · PASCAL VOC Novel Split 1, 1-shot | nAP50: 41.3 -> 38.9; 41.3 -> 37.9 | Hurt |
| Restrict CosFace or ArcFace to novel samples (ablation) | TFA · PASCAL VOC Novel Split 1, 1-shot | nAP50: 38.9 -> 44.2; 37.9 -> 44.3 | Helped |
| Use WordNet semantic similarity rather than visual similarity for assignments (ablation) | FADI association · PASCAL VOC Novel Split 1, 1-shot | nAP50: 43.3 -> 44.9 | Helped |
| Use top-1 semantic assignment rather than random assignment (ablation) | FADI association without margin loss · PASCAL VOC Novel Split 1 | nAP50: 39.6 -> 44.3 | Helped |
Closest papers by meaning
- Frustratingly Simple Few-Shot Object Detection (ICML 2020 · similarity 0.90)
- DiGeo: Discriminative Geometry-Aware Learning for Generalized Few-Shot Object Detection (CVPR 2023 · similarity 0.90)
- A Closer Look at Cross-Domain Few-Shot Object Detection: Fine-Tuning Matters and Parallel Decoder Helps (CVPR 2026 · similarity 0.90)
- Fine-Grained Prototypes Distillation for Few-Shot Object Detection (AAAI 2024 · similarity 0.90)
- Few-Shot Object Detection via Classification Refinement and Distractor Retreatment (CVPR 2021 · similarity 0.89)
- Few-Shot Object Detection via Variational Feature Aggregation (AAAI 2023 · 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.