Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
F-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning
Check 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 |
|---|---|---|---|
| Combine VAE and GAN losses in the feature generator (ablation) | f-VAEGAN · CUB inductive ZSL | top-1 accuracy: 59.1% and 58.4% vs 61.0% | Helped |
| Combine VAE and GAN losses in the feature generator (ablation) | f-VAEGAN · CUB inductive GZSL | harmonic mean: 52.3 and 52.5 vs 53.7 | Helped |
| Combine VAE and GAN losses in the feature generator (ablation) | f-VAEGAN · CUB transductive ZSL | top-1 accuracy: 67.3 and 68.9 vs 71.1 | Helped |
| Combine VAE and GAN losses in the feature generator (ablation) | f-VAEGAN · CUB transductive GZSL | harmonic mean: 59.6 and 61.6 vs 63.2 | Helped |
| Add unlabeled samples during training (ablation) | GAN, VAE, and VAE-GAN feature generators · CUB ZSL | top-1 accuracy | Helped |
| Fine-tune image features on seen-class images | f-VAEGAN · Inductive and transductive ZSL | top-1 accuracy: 61.0 -> 72.9; 71.1 -> 82.6 (CUB) | Helped |
| Fine-tune image features on seen-class images | f-VAEGAN · AWA transductive ZSL | top-1 accuracy: 89.8 -> 89.3 | Hurt |
| Use feature generation for generalized zero-shot classification | feature-generating methods · GZSL | harmonic mean | Helped |
| Use unlabeled novel-class feature data with the non-conditional discriminator D2 | f-VAEGAN-D2 · Transductive any-shot learning | feature generation quality | Helped |
Closest papers by meaning
- Feature Generating Networks for Zero-Shot Learning (CVPR 2018 · similarity 0.91)
- Zero-Shot Learning via Class-Conditioned Deep Generative Models (AAAI 2018 · similarity 0.90)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 2025 · similarity 0.87)
- Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition (AAAI 2024 · similarity 0.87)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.86)
- Data-Free Generalized Zero-Shot Learning (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.