Research map / Image generation / GANs
An Adversarial Framework for Generating Unseen Images by Activation Maximization
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 PROBEGAN instead of BIGGAN-AM for unseen-class generation | CIFAR-10 unseen-class generation · CIFAR-10, w/o horse | intra-FID (horse): 223.3 -> 45.40 | Helped |
| Use PROBEGAN instead of BIGGAN-AM for unseen-class generation | CIFAR-10 unseen-class generation · CIFAR-10, w/o truck | intra-FID (truck): 161.5 -> 68.33 | Helped |
| Use a robust rather than regular classifier | PROBEGAN · CIFAR-10, w/o horse | intra-FID (horse): 91.35 -> 45.40 | Helped |
| Use a robust rather than regular classifier | PROBEGAN · CIFAR-10, w/o truck | intra-FID (truck): 105.99 -> 68.33 | Helped |
| Remove the conditional discriminator and class dilution (ablation) | PROBEGAN (NAIVE variant) · CIFAR-10, w/o horse | intra-FID (horse): 45.40 -> 76.02 | Hurt |
| Remove the conditional discriminator and class dilution (ablation) | PROBEGAN (NAIVE variant) · CIFAR-10, w/o truck | intra-FID (truck): 68.33 -> 84.21 | Hurt |
| Include target-class images in the training set, with an artificial marker (ablation) | PROBEGAN · CIFAR-10 truck generation | presence of the artificial red blocks in generated images | Helped |
Closest papers by meaning
- Activation Maximization Generative Adversarial Nets (ICLR 2018 · similarity 0.90)
- RareGAN: Generating Samples for Rare Classes (AAAI 2022 · similarity 0.88)
- Collapse by Conditioning: Training Class-conditional GANs with Limited Data (ICLR 2022 · similarity 0.87)
- SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer (ICLR 2024 · similarity 0.87)
- Taming the Tail in Class-Conditional GANs: Knowledge Sharing via Unconditional Training at Lower Resolutions (CVPR 2024 · similarity 0.86)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · 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.