Research map / Image generation / GANs
DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data
Check 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 the fixed latent prior with a trainable mixture of Gaussians | DCGAN · CIFAR-10 (2000 images) | m-IS overall: 2.15 ± 0.25 -> 2.28 ± 0.62 | Helped |
| Replace the fixed latent prior with a trainable mixture of Gaussians | DCGAN · TU-Berlin sketches, similar categories | m-IS overall: 1.36 ± 0.09 -> 1.35 ± 0.10 | No clear effect |
| Replace the fixed latent prior with a trainable mixture of Gaussians | DCGAN · TU-Berlin sketches, dissimilar categories | m-IS overall: 1.61 ± 0.13 -> 1.79 ± 0.18 | Helped |
| Replace the fixed latent prior with a trainable mixture of Gaussians | DCGAN · CIFAR-10, Cat and Dog classes | m-IS: Cat: 2.43 ± 0.19 -> 2.17 ± 0.04; Dog: 2.22 ± 0.23 -> 1.27 ± 0.01 | Mixed |
| Use a trainable mixture of Gaussians instead of a one-hot categorical latent variable | CIFAR-10 GAN · CIFAR-10 (2000 images) | m-IS overall: 2.04 ± 0.28 -> 2.28 ± 0.62 | Helped |
| Use a trainable mixture of Gaussians instead of adding a fully connected layer | CIFAR-10 GAN · CIFAR-10 (2000 images) | m-IS overall: 1.76 ± 0.40 -> 2.28 ± 0.62 | Helped |
Closest papers by meaning
- AdaGAN: Boosting Generative Models (NeurIPS 2017 · similarity 0.91)
- MGAN: Training Generative Adversarial Nets with Multiple Generators (ICLR 2018 · similarity 0.91)
- Constrained Generative Adversarial Networks for Interactive Image Generation (CVPR 2019 · similarity 0.90)
- Training Generative Adversarial Networks with Limited Data (NeurIPS 2020 · similarity 0.90)
- Large Scale GAN Training for High Fidelity Natural Image Synthesis (ICLR 2019 · similarity 0.90)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.90)
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.