Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Feature Distribution Fitting with Direction-Driven Weighting for Few-Shot Images Classification
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 DDWM feature-distribution fitting | Few-shot image classification · miniImageNet 5-way-1-shot | accuracy: 68.32 -> 68.58 | Helped |
| Use DDWM feature-distribution fitting | Few-shot image classification · miniImageNet 5-way-5-shot | accuracy: 83.18 -> 84.65 | Helped |
| Use DDWM feature-distribution fitting | Few-shot image classification · CUB 5-way-1-shot | accuracy: 79.60 -> 80.40 | Helped |
| Use DDWM feature-distribution fitting | Few-shot image classification · CUB 5-way-5-shot | accuracy: 90.48 -> 90.75 | Helped |
| Use DDWM feature-distribution fitting | Few-shot image classification · CIFAR-FS 5-way-1-shot | accuracy: 74.51 -> 75.26 | Helped |
| Use DDWM feature-distribution fitting | Few-shot image classification · CIFAR-FS 5-way-5-shot | accuracy: 86.60 -> 86.90 | Helped |
| Add Tukey transformation only (ablation) | DDWM on miniImageNet · miniImageNet 5-way-1-shot | accuracy: 60.06 -> 64.66 | Helped |
| Add generated features only (ablation) | DDWM on miniImageNet · miniImageNet 5-way-1-shot | accuracy: 60.06 -> 64.93 | Helped |
| Add L2 normalization only (ablation) | DDWM on miniImageNet · miniImageNet 5-way-1-shot | accuracy: 60.06 -> 66.64 | Helped |
| Add Tukey transformation and generated features (ablation) | DDWM on miniImageNet · miniImageNet 5-way-1-shot | accuracy: 60.06 -> 68.14 | Helped |
| Add Tukey transformation, generated features, and L2 normalization (ablation) | DDWM on miniImageNet · miniImageNet 5-way-1-shot | accuracy: 60.06 -> 68.58 | Helped |
| Increase the number of generated features (ablation) | DDWM on miniImageNet · miniImageNet 5-way-1-shot | accuracy | Helped |
Closest papers by meaning
- Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformation (ICLR 2020 · similarity 0.88)
- Improved Few-Shot Visual Classification (CVPR 2020 · similarity 0.88)
- Few-Shot Classification With Feature Map Reconstruction Networks (CVPR 2021 · similarity 0.88)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 2024 · similarity 0.86)
- Provably Improving Generalization of Few-shot models with Synthetic Data (ICML 2025 · similarity 0.86)
- FeatWalk: Enhancing Few-Shot Classification through Local View Leveraging (AAAI 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.