Research map / Detection and segmentation / Open-set and few-shot detection
Differentiable Meta-Learning Model for Few-Shot Semantic Segmentation
Pinzhuo Tian, Zhangkai Wu, Lei Qi, Lei Wang, Yinghuan Shi, Yang Gao
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 pixel classification with a differentiable linear base learner for multi-object segmentation | MetaSegNet · PASCAL-5^2, 2-way 5-shot | Mean IoU (%): 42.6 -> 43.3 | Improved |
| Use MetaSegNet instead of the strongest listed baseline | COCO, 2-way few-shot segmentation · COCO, 2-way 1-shot | Mean IoU (%): 29.3 -> 33.2 | Improved |
| Increase support examples from 1-shot to 10-shot | MetaSegNet · COCO, 2-way segmentation | Mean IoU (%): 33.2 -> 38.3 | Improved |
| Use MetaSegNet with miniImageNet pretraining instead of OSLSM | PASCAL-5^i, 1-way 5-shot · PASCAL-5^i | Mean IoU (%): 61.5 -> 59.5 | Hurt |
| Add episodic meta-training rather than fine-tuning (ablation) | Few-shot segmentation on PASCAL-5^i_val · PASCAL-5^i_val, 2-way 1-shot | Mean IoU (%): 27.28 -> 35.00 | Improved |
| Add ridge regression as the linear base learner instead of a 1×1 convolution (ablation) | MetaSegNet-NG · PASCAL-5^i_val, 2-way 1-shot | Mean IoU (%): 35.00 -> 37.75 | Improved |
| Add the global context branch (ablation) | MetaSegNet · PASCAL-5^i_val, 2-way 1-shot | Mean IoU (%): 37.75 -> 38.92 | Improved |
| Add ridge regression as the linear base learner instead of a 1×1 convolution (ablation) | MetaSegNet-NG · PASCAL-5^i_val, 2-way 5-shot | Mean IoU (%): 36.04 -> 48.34 | Improved |
| Add the global context branch (ablation) | MetaSegNet · PASCAL-5^i_val, 2-way 5-shot | Mean IoU (%): 48.34 -> 49.44 | Improved |
| Increase support examples from 1-shot to 5-shot (ablation) | MetaSegNet · PASCAL-5^i_val, 2-way segmentation | Mean IoU (%) | Improved |
| Train MetaSegNet from scratch rather than pretraining on miniImageNet | MetaSegNet · PASCAL-5^i, 1-way 5-shot | Mean IoU (%): 59.5 -> 53.4 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PANet — PANet: Few-Shot Image Semantic Segmentation With Prototype Alignment (ICCV 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning Meta-Class Memory for Few-Shot Semantic Segmentation (ICCV 2021 · similarity 0.90)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.89)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.88)
- APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation (CVPR 2024 · similarity 0.87)
- Unleashing the Potential of the Diffusion Model in Few-shot Semantic Segmentation (NeurIPS 2024 · similarity 0.86)
- A Surprisingly Simple Approach to Generalized Few-Shot Semantic Segmentation (NeurIPS 2024 · similarity 0.86)
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.
