Research map / Image generation / GANs
Attribute Group Editing for Reliable 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 |
|---|---|---|---|
| Use AGE instead of the strongest reported baseline | Few-shot image generation · Flowers | FID: 79.33 -> 45.96 | Helped |
| Use AGE instead of the strongest reported baseline | Few-shot image generation · Flowers | LPIPS: 0.3912 -> 0.4305 | Helped |
| Use AGE instead of the strongest reported baseline | Few-shot image generation · Animal Faces | FID: 89.81 -> 28.04 | Helped |
| Use AGE instead of the strongest reported baseline | Few-shot image generation · Animal Faces | LPIPS: 0.4964 -> 0.5575 | Helped |
| Use AGE instead of the strongest reported baseline | Few-shot image generation · VGG Faces | LPIPS: 0.3146 -> 0.3294 | Helped |
| Use AGE instead of LoFGAN | Few-shot image generation · VGG Faces | FID: 20.31 -> 34.86 | Hurt |
| Factorize category-irrelevant edits with SDL rather than directly sampling training latent differences (ablation) | AGE data augmentation · Animal Faces classification | FID: 54.66 -> 38.18 | Helped |
| Set manipulation intensity α to 1.0 rather than 0.3 (ablation) | AGE data augmentation · Animal Faces classification | Accuracy: 69.8 -> 71.4 | Helped |
| Set manipulation intensity α to 1.0 rather than 2.0 (ablation) | AGE data augmentation · Animal Faces classification | Accuracy: 66.1 -> 71.4 | Helped |
| Increase manipulation intensity from α=0.3 to α=2.0 (ablation) | AGE image generation · Animal Faces | LPIPS: 0.5429 -> 0.5809 | Helped |
| Increase manipulation intensity from α=0.3 to α=2.0 (ablation) | AGE image generation · Animal Faces | FID: 31.79 -> 63.99 | Hurt |
| Add the orthogonality loss L_orth (ablation) | AGE few-shot image classification · Animal Faces | Accuracy: 69.8 -> 71.4 | Helped |
Closest papers by meaning
- Enjoy Your Editing: Controllable GANs for Image Editing via Latent Space Navigation (ICLR 2021 · similarity 0.87)
- EditGAN: High-Precision Semantic Image Editing (NeurIPS 2021 · similarity 0.87)
- LARGE: Latent-Based Regression Through GAN Semantics (CVPR 2022 · similarity 0.87)
- Low-Rank Subspaces in GANs (NeurIPS 2021 · similarity 0.87)
- Generative Attribute Controller With Conditional Filtered Generative Adversarial Networks (CVPR 2017 · similarity 0.87)
- E$^2$GAN: Efficient Training of Efficient GANs for Image-to-Image Translation (ICML 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.