Research map / Image generation / GANs
cGANs with Projection Discriminator
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 |
|---|---|---|---|
| replace hidden concatenation with a projection discriminator | class-conditional GAN · ImageNet (ILSVRC2012) | Inception Score: 21.1$\pm$.35 -> 29.7$\pm$.61 | Helped |
| replace hidden concatenation with a projection discriminator | class-conditional GAN · ImageNet (ILSVRC2012) | Intra FID: 141.2 -> 103.1 | Helped |
| train the projection model for 850K rather than 450K iterations (ablation) | Projection · ImageNet (ILSVRC2012) | Inception Score: 29.7$\pm$.61 -> 36.8$\pm$.44 | Helped |
| train the projection model for 850K rather than 450K iterations (ablation) | Projection · ImageNet (ILSVRC2012) | Intra FID: 103.1 -> 92.4 | Helped |
| use a projection discriminator rather than concatenation | super-resolution GAN · ImageNet super-resolution validation | Inception accuracy (%): 11.0 -> 35.2 | Helped |
| use a projection discriminator rather than concatenation | super-resolution GAN · ImageNet super-resolution validation | MS-SSIM: 0.829 -> 0.878 | Helped |
| average logits over 10 random seeds (ablation) | Projection super-resolution model · ImageNet super-resolution validation | Inception accuracy (%): 35.2 -> 36.4 | Helped |
| use a projection discriminator rather than hidden concatenation | class-conditional GAN · CIFAR-10 | Inception score: 8.14 -> 8.62 | Helped |
| use a projection discriminator rather than hidden concatenation | class-conditional GAN · CIFAR-10 | FID: 19.2 -> 17.5 | Helped |
| use a projection discriminator rather than hidden concatenation | class-conditional GAN · CIFAR-100 | Inception score: 8.82 -> 9.04 | Helped |
| use a projection discriminator rather than hidden concatenation | class-conditional GAN · CIFAR-100 | FID: 24.8 -> 23.2 | Helped |
| add pretrained auxiliary-classifier loss during the final 50K updates (ablation) | Projection · ImageNet (ILSVRC2012) | Inception Score | Helped |
Closest papers by meaning
- A Unified View of cGANs with and without Classifiers (NeurIPS 2021 · similarity 0.89)
- CcGAN: Continuous Conditional Generative Adversarial Networks for Image Generation (ICLR 2021 · similarity 0.89)
- Conditional Image Synthesis with Auxiliary Classifier GANs (ICML 2017 · similarity 0.89)
- Robust Conditional Generative Adversarial Networks (ICLR 2019 · similarity 0.88)
- SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer (ICLR 2024 · similarity 0.86)
- Turning Waste into Wealth: Leveraging Low-Quality Samples for Enhancing Continuous Conditional Generative Adversarial Networks (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.