Research map / Detection and segmentation / Open-set and few-shot detection
A Strong Baseline for Generalized Few-Shot Semantic Segmentation
Sina Hajimiri, Malik Boudiaf, Ismail Ben Ayed, Jose Dolz
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 |
|---|---|---|---|
| Use DIaM instead of BAM | GFSS with ResNet-50 · PASCAL-5i 1-shot | Novel mIoU: 27.49 -> 35.11 | Improved |
| Use DIaM instead of BAM | GFSS with ResNet-50 · PASCAL-5i 5-shot | Novel mIoU: 28.96 -> 55.31 | Improved |
| Use DIaM instead of BAM | GFSS with ResNet-50 · COCO-20i 1-shot | Novel mIoU: 14.16 -> 17.22 | Improved |
| Use DIaM instead of BAM | GFSS with ResNet-50 · COCO-20i 5-shot | Novel mIoU: 16.63 -> 28.73 | Improved |
| Use DIaM instead of BAM | GFSS with ResNet-50 · PASCAL-10i 1-shot | Novel mIoU: 15.48 -> 31.29 | Improved |
| Use DIaM instead of BAM | GFSS with ResNet-50 · PASCAL-10i 5-shot | Novel mIoU: 21.51 -> 51.89 | Improved |
| Use a more accurate true class-proportion prior instead of the estimated prior (ablation) | DIaM · PASCAL-5i 5-shot | Novel mIoU: 55.31 -> 66.12 | Improved |
| Use a more accurate true class-proportion prior instead of the estimated prior (ablation) | DIaM · COCO-20i 5-shot | Novel mIoU: 28.73 -> 40.43 | Improved |
| Estimate the class-proportion prior from model predictions rather than use a uniform prior (ablation) | DIaM · PASCAL-5i 5-shot | Novel mIoU | Improved |
| Add the knowledge-distillation term to preserve base-class predictions (ablation) | DIaM · PASCAL-5i | Base-class preservation | Improved |
| Optimize the base classifier with distillation instead of freezing it (ablation) | DIaM · PASCAL-5i 5-shot | Convergence and optimum | Improved |
| Increase adaptation from 10 to 100 iterations (ablation) | DIaM · PASCAL-5i 5-shot | Mean mIoU: 59.22 -> 63.08 | Improved |
Closest papers by meaning
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.94)
- Generalized Few-Shot Semantic Segmentation (CVPR 2022 · similarity 0.93)
- Enhancing Generalized Few-Shot Semantic Segmentation via Effective Knowledge Transfer (AAAI 2025 · similarity 0.91)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.91)
- Prototypical Kernel Learning and Open-set Foreground Perception for Generalized Few-shot Semantic Segmentation (ICCV 2023 · similarity 0.90)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.89)
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.
