# An Image-based Generator Architecture for Synthetic Image Refinement

arXiv: https://arxiv.org/abs/2108.04957 · HTML: https://mlsanity.com/paper/2108.04957/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 13 references that are in the map.
- First posted on arXiv 10 Aug 2021.

## Builds on (most cited first, 11 of 11)

- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Learning from Simulated and Unsupervised Images through Adversarial Training](https://mlsanity.com/paper/1612.07828/) (2016)
- [Deep Appearance Models for Face Rendering](https://mlsanity.com/paper/1808.00362/) (2018)
- [BEGAN: Boundary Equilibrium Generative Adversarial Networks](https://mlsanity.com/paper/1703.10717/) (2017)
- [Energy-based Generative Adversarial Network](https://mlsanity.com/paper/1609.03126/) (2016)
- [AvatarMe: Realistically Renderable 3D Facial Reconstruction "in-the-wild"](https://mlsanity.com/paper/2003.13845/) (2020)
- [Photo-Realistic Facial Details Synthesis from Single Image](https://mlsanity.com/paper/1903.10873/) (2019)
- [LT-GAN: Self-Supervised GAN with Latent Transformation Detection](https://mlsanity.com/paper/2010.09893/) (2020)
- [Finding your Lookalike: Measuring Face Similarity Rather than Face Identity](https://mlsanity.com/paper/1806.05252/) (2018)

## Abstract (arXiv)

Proposed are alternative generator architectures for Boundary Equilibrium Generative Adversarial Networks, motivated by Learning from Simulated and Unsupervised Images through Adversarial Training. It disentangles the need for a noise-based latent space. The generator will operate mainly as a refiner network to gain a photo-realistic presentation of the given synthetic images. It also attempts to resolve the latent space's poorly understood properties by eliminating the need for noise injection and replacing it with an image-based concept. The new flexible and simple generator architecture will also give the power to control the trade-off between restrictive refinement and expressiveness ability. Contrary to other available methods, this architecture will not require a paired or unpaired dataset of real and synthetic images for the training phase. Only a relatively small set of real images would suffice.