Research map / Detection and segmentation / Open-set and few-shot detection
One-Shot Unsupervised Cross-Domain Detection
Antonio D’Innocente, Francesco Cappio Borlino, Silvia Bucci, Barbara Caputo, Tatiana Tommasi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 pseudo-box-crop rotation task instead of whole-image rotation (ablation) | OSHOT (gamma=30) · VOC -> Clipart | mAP: 31.0 -> 33.9 | Improved |
| add pseudo-box-crop rotation task instead of whole-image rotation (ablation) | OSHOT (gamma=30) · VOC -> Comic | mAP: 21.0 -> 26.9 | Improved |
| add pseudo-box-crop rotation task instead of whole-image rotation (ablation) | OSHOT (gamma=30) · VOC -> Watercolor | mAP: 48.2 -> 52.0 | Improved |
| add pseudo-box-crop rotation task instead of whole-image rotation (ablation) | OSHOT (gamma=30) · Cityscapes -> FoggyCityscapes | mAP: 27.7 -> 31.9 | Improved |
| add 30 one-shot self-supervised adaptation iterations | OSHOT (gamma=0) · VOC -> Clipart | mAP: 28.3 -> 33.9 | Improved |
| add 30 one-shot adaptation iterations | OSHOT (gamma=0) · VOC -> Comic | mAP: 18.8 -> 26.9 | Improved |
| add 30 one-shot adaptation iterations | OSHOT (gamma=0) · VOC -> Watercolor | mAP: 46.1 -> 52.0 | Improved |
| add 30 one-shot adaptation iterations | OSHOT (gamma=0) · Cityscapes -> FoggyCityscapes | mAP: 29.3 -> 31.9 | Improved |
| add 30 one-shot adaptation iterations | OSHOT (gamma=0) · KITTI -> Cityscapes (car) | mAP: 26.2 -> 33.5 | Improved |
| add 30 one-shot adaptation iterations | OSHOT (gamma=0) · Cityscapes -> KITTI (car) | mAP: 75.4 -> 75.0 | No clear effect |
| add self-supervised rotation task at training time only (gamma=0) | FRCNN · VOC -> Social Bikes | mAP: 62.1 -> 62.4 | No clear effect |
| use one-shot style transfer (BiOST) before inference | FRCNN · Social Bikes | mAP: 62.1 -> 51.1 | Hurt |
Closest papers by meaning
- Cross-Domain Adaptive Teacher for Object Detection (CVPR 2022 · similarity 0.91)
- Cross Domain Object Detection by Target-Perceived Dual Branch Distillation (CVPR 2022 · similarity 0.90)
- MTTrans: Cross-Domain Object Detection with Mean Teacher Transformer (ECCV 2022 · similarity 0.90)
- Cross-Domain Weakly-Supervised Object Detection Through Progressive Domain Adaptation (CVPR 2018 · similarity 0.90)
- AcroFOD: An Adaptive Method for Cross-Domain Few-Shot Object Detection (ECCV 2022 · similarity 0.90)
- Unbiased Mean Teacher for Cross-Domain Object Detection (CVPR 2021 · similarity 0.90)
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.
