Research map / Image generation / GANs
Smooth Deep Image Generator from Noises
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 |
|---|---|---|---|
| Set Lipschitz-type constraint k to 10 instead of 100 (ablation) | Smooth GAN · CIFAR-10 | FID: 26.3 -> 24.1 | Helped |
| Set k to 10 instead of 3 (ablation) | Smooth GAN · CIFAR-10 | FID: 28.6 -> 24.1 | Helped |
| Set k to 10 instead of 0.1 (ablation) | Smooth GAN · CIFAR-10 | FID: 46.1 -> 24.1 | Helped |
| Use Smooth GAN instead of WGAN-GP | Smooth GAN · CIFAR-10 | FID: 26.5 -> 24.1 | Helped |
| Use Smooth GAN instead of WGAN-GP | Smooth GAN · CIFAR-10 | Inception score: 7.36 -> 7.66 | Helped |
| Use Smooth GAN instead of D2GAN | Smooth GAN · CIFAR-10 | FID: 30.6 -> 24.1 | Helped |
| Use Smooth GAN instead of DCGAN | Smooth GAN · CIFAR-10 | Inception score: 6.40 -> 7.66 | Helped |
| Use Smooth GAN instead of WGAN-GP for interpolation smoothness | Smooth GAN · CelebA interpolation slices | sliced MS-SSIM and FID | Helped |
| Use Smooth GAN instead of WGAN-GP on MNIST interpolation | Smooth GAN · MNIST interpolation | qualitative meaningful images | Helped |
| Use Smooth GAN instead of DCGAN and WGAN-GP in subset FID distribution | Smooth GAN · CIFAR-10 FID over 120 subsets of 512 images | FID distribution | Helped |
Closest papers by meaning
- Noise Robust Generative Adversarial Networks (CVPR 2020 · similarity 0.90)
- Smoothness and Stability in GANs (ICLR 2020 · similarity 0.89)
- Progressive Growing of GANs for Improved Quality, Stability, and Variation (ICLR 2018 · similarity 0.88)
- Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training (NeurIPS 2021 · similarity 0.87)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.86)
- Consistency-GAN: Training GANs with Consistency Model (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.