Research map / Image generation / GANs
Conditional Image Synthesis with Auxiliary Classifier GANs
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 AC-GAN instead of the prior CIFAR-10 state-of-the-art model | CIFAR-10 image synthesis · CIFAR-10 | Inception score: 8.09 ± 0.07 -> 8.25 ± 0.07 | Helped |
| Generate at 128×128 rather than resizing outputs to 64×64 (ablation) | AC-GAN · ImageNet, 1000 classes | Inception accuracy: 7.0% ± 2.0% -> 10.1% ± 2.0% | Helped |
| Generate at 128×128 rather than resizing outputs to 32×32 (ablation) | AC-GAN · ImageNet, 1000 classes | Inception accuracy: 5.0% ± 2.0% -> 10.1% ± 2.0% | Helped |
| Train at 128×128 instead of 64×64 (ablation) | AC-GAN · ImageNet, 1000 classes | Inception accuracy | Helped |
| Increase the number of classes per training model (ablation) | AC-GAN · ImageNet class splits | Sample quality | Hurt |
| Increase model parameters when training on larger class splits (ablation) | AC-GAN · ImageNet class splits | Sample quality | Helped |
| Use a small class split rather than one class per model (ablation) | AC-GAN · ImageNet class splits | Training convergence | Helped |
| Change which classes are grouped together in a split (ablation) | AC-GAN · ImageNet class splits | Sample quality | No clear effect |
Closest papers by meaning
- Large Scale GAN Training for High Fidelity Natural Image Synthesis (ICLR 2019 · similarity 0.93)
- Diffusion Models Beat GANs on Image Synthesis (NeurIPS 2021 · similarity 0.91)
- Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis (ICLR 2021 · similarity 0.90)
- Self-Supervised GANs via Auxiliary Rotation Loss (CVPR 2019 · similarity 0.90)
- Taming the Tail in Class-Conditional GANs: Knowledge Sharing via Unconditional Training at Lower Resolutions (CVPR 2024 · similarity 0.89)
- Unlocking Pre-trained Image Backbones for Semantic Image Synthesis (CVPR 2024 · similarity 0.89)
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.