Research map / Detection and segmentation / Open-set and few-shot detection
Incremental-DETR: Incremental Few-Shot Object Detection via Self-Supervised Learning
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 Incremental-DETR rather than iMTFA | incremental few-shot detection · COCO val, 5-shot, all classes | AP: 19.6 -> 24.9 | Helped |
| Use Incremental-DETR rather than iMTFA | incremental few-shot detection · COCO val, 10-shot, all classes | AP: 19.3 -> 24.1 | Helped |
| Use Incremental-DETR rather than ONCE | incremental few-shot detection · VOC test, 5-shot, novel classes | AP: 2.4 -> 16.6 | Helped |
| Add the two-stage fine-tuning, self-supervised losses, and distillation components (ablation) | Incremental-DETR · COCO val, 5-shot, all classes | AP: 0.4 -> 24.9 | Helped |
| Add the self-supervised box and classification losses (ablation) | Incremental-DETR with two-stage fine-tuning · COCO val, 5-shot, novel classes | AP: 5.2 -> 8.0 | Helped |
| Add feature distillation (ablation) | Incremental-DETR with self-supervised losses and classification distillation · COCO val, 5-shot, all classes | AP: 21.6 -> 24.9 | Helped |
| Add classification distillation (ablation) | Incremental-DETR with self-supervised losses and feature distillation · COCO val, 5-shot, all classes | AP: 19.9 -> 24.9 | Helped |
| Use two-step rather than one-step base pre-training (ablation) | Incremental-DETR · COCO val, 5-shot, novel classes | AP: 7.1 -> 8.3 | Helped |
| Keep class-agnostic modules frozen during novel-class fine-tuning (ablation) | Incremental-DETR · COCO val, 5-shot, all classes | AP: 4.1 -> 24.9 | Helped |
Closest papers by meaning
- Incremental Few-Shot Object Detection (CVPR 2020 · similarity 0.93)
- Frustratingly Simple Few-Shot Object Detection (ICML 2020 · similarity 0.90)
- Sylph: A Hypernetwork Framework for Incremental Few-Shot Object Detection (CVPR 2022 · similarity 0.90)
- Label, Verify, Correct: A Simple Few Shot Object Detection Method (CVPR 2022 · similarity 0.89)
- MS-DETR: Efficient DETR Training with Mixed Supervision (CVPR 2024 · similarity 0.88)
- Weakly Supervised Few-Shot Object Detection with DETR (AAAI 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.