Research map / Image generation / GANs
A Closer Look at Few-Shot Image Generation
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 |
|---|---|---|---|
| Add dual contrastive learning to few-shot GAN adaptation | StyleGAN2 · FFHQ → Babies, 10-shot | FID: 74.39 ± 0.03 -> 52.56 ± 0.02 | Helped |
| Add dual contrastive learning to few-shot GAN adaptation | StyleGAN2 · FFHQ → Sunglasses, 10-shot | FID: 42.13 ± 0.04 -> 38.01± 0.01 | Helped |
| Add dual contrastive learning to few-shot GAN adaptation | StyleGAN2 · FFHQ → Sketches, 10-shot | FID: 45.67 ± 0.02 -> 37.90 ± 0.02 | Helped |
| Add dual contrastive learning to few-shot GAN adaptation | StyleGAN2 · FFHQ → Otto's Paintings, 10-shot | intra-LPIPS: 0.63 ± 0.03 -> 0.66 ± 0.02 | Helped |
| Add dual contrastive learning to few-shot GAN adaptation | StyleGAN2 · Church → Haunted House, 10-shot | intra-LPIPS: 0.60 ± 0.04 -> 0.63 ± 0.01 | Helped |
| Add dual contrastive learning to few-shot GAN adaptation | StyleGAN2 · Cars → Abandoned Cars, 10-shot | intra-LPIPS: 0.52 ± 0.04 -> 0.53 ± 0.02 | Helped |
| Add Generator CL (ablation) | DCL · Few-shot adaptation; diversity during training | diversity | Helped |
| Add Discriminator CL (ablation) | DCL · Few-shot adaptation; diversity during training | diversity | Helped |
| Combine Generator CL and Discriminator CL (ablation) | DCL · Few-shot adaptation; diversity during training | diversity | Helped |
| Use target-generator outputs rather than source-generator outputs as Generator CL negatives (ablation) | DCL · FFHQ → Amedeo's paintings, 10-shot | intra-LPIPS | No clear effect |
| Use generated images rather than real target images as Discriminator CL negatives (ablation) | DCL · Few-shot adaptation | diversity | No clear effect |
| Train the realisticness classifier with alternative non-target datasets (ablation) | Binary realisticness classifier · FFHQ → Sketches adaptation | target-domain realisticness | No clear effect |
Closest papers by meaning
- Few-shot Image Generation via Adaptation-Aware Kernel Modulation (NeurIPS 2022 · similarity 0.91)
- Few-Shot Image Generation via Cross-Domain Correspondence (CVPR 2021 · similarity 0.89)
- Diverse Image Generation via Self-Conditioned GANs (CVPR 2020 · similarity 0.89)
- Diverse Rare Sample Generation with Pretrained GANs (AAAI 2025 · similarity 0.89)
- Exact Fusion via Feature Distribution Matching for Few-shot Image Generation (CVPR 2024 · similarity 0.88)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.87)
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.