Research map / Image generation / GANs
Training Generative Adversarial Networks from Incomplete Observations using Factorised Discriminators
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 FactorGAN with unpaired observations in addition to paired examples | Cityscapes segmentation · Cityscapes, 25 paired examples | pixel-wise accuracy: 14.9% over CycleGAN; 17.7% over GAN | Helped |
| Use FactorGAN instead of a standard GAN when additional unpaired examples are available | Image pair generation · Edges2Shoes | LS metric | Helped |
| Use FactorGAN instead of a standard GAN | Image pair generation · Cityscapes | LS metric | Helped |
| Use FactorGAN instead of a standard GAN when additional incomplete examples are available | Audio source separation · MUSDB | SDR | Helped |
| Increase the number of paired observations | FactorGAN · Paired MNIST, digit dependency metric | dependency metric | Helped |
| Remove the p-dependency discriminator (ablation) | FactorGAN · Paired MNIST, λ=0.9 | dependency metric | Hurt |
| Remove the p-dependency discriminator (ablation) | FactorGAN · Paired MNIST, λ=0.1 | dependency metric | No clear effect |
| Increase the number of paired observations | GAN · Audio source separation | SDR | Hurt |
| Use CycleGAN rather than GAN | Cityscapes segmentation · Cityscapes | pixel-wise accuracy | Mixed |
Closest papers by meaning
- MGAN: Training Generative Adversarial Nets with Multiple Generators (ICLR 2018 · similarity 0.91)
- AdaGAN: Boosting Generative Models (NeurIPS 2017 · similarity 0.90)
- Training Generative Adversarial Networks with Limited Data (NeurIPS 2020 · similarity 0.90)
- Mixed batches and symmetric discriminators for GAN training (ICML 2018 · similarity 0.90)
- Training Generative Adversarial Networks in One Stage (CVPR 2021 · similarity 0.90)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.88)
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.