# Memory Efficient Patch-based Training for INR-based GANs

arXiv: https://arxiv.org/abs/2207.01395 · HTML: https://mlsanity.com/paper/2207.01395/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 27 references that are in the map.
- First posted on arXiv 4 Jul 2022.

## Builds on (most cited first, 25 of 25)

- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation](https://mlsanity.com/paper/1901.05103/) (2019)
- [Implicit Neural Representations with Periodic Activation Functions](https://mlsanity.com/paper/2006.09661/) (2020)
- [StarGAN v2: Diverse Image Synthesis for Multiple Domains](https://mlsanity.com/paper/1912.01865/) (2019)
- [StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation](https://mlsanity.com/paper/1711.09020/) (2017)
- [GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields](https://mlsanity.com/paper/2011.12100/) (2020)
- [pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis](https://mlsanity.com/paper/2012.00926/) (2020)
- [Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?](https://mlsanity.com/paper/1904.03189/) (2019)
- [StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis](https://mlsanity.com/paper/2110.08985/) (2021)
- [In-Domain GAN Inversion for Real Image Editing](https://mlsanity.com/paper/2004.00049/) (2020)
- [SAL: Sign Agnostic Learning of Shapes from Raw Data](https://mlsanity.com/paper/1911.10414/) (2019)
- [StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generation](https://mlsanity.com/paper/2112.11427/) (2021)
- [Learning Shape Templates with Structured Implicit Functions](https://mlsanity.com/paper/1904.06447/) (2019)
- [U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation](https://mlsanity.com/paper/1907.10830/) (2019)
- [Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks](https://mlsanity.com/paper/2202.10571/) (2022)
- [Adversarial Generation of Continuous Images](https://mlsanity.com/paper/2011.12026/) (2020)
- [Image Generators with Conditionally-Independent Pixel Synthesis](https://mlsanity.com/paper/2011.13775/) (2020)
- [Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editing](https://mlsanity.com/paper/2104.14754/) (2021)
- [NSML: Meet the MLaaS platform with a real-world case study](https://mlsanity.com/paper/1810.09957/) (2018)
- [COCO-GAN: Generation by Parts via Conditional Coordinating](https://mlsanity.com/paper/1904.00284/) (2019)
- [Tag2Pix: Line Art Colorization Using Text Tag With SECat and Changing Loss](https://mlsanity.com/paper/1908.05840/) (2019)
- [NSML: A Machine Learning Platform That Enables You to Focus on Your Models](https://mlsanity.com/paper/1712.05902/) (2017)

## Built on by (most cited first, 2 of 2)

- [Patch Diffusion: Faster and More Data-Efficient Training of Diffusion Models](https://mlsanity.com/paper/2304.12526/) (2023)
- [Efficient-3DiM: Learning a Generalizable Single-image Novel-view Synthesizer in One Day](https://mlsanity.com/paper/2310.03015/) (2023)

## Abstract (arXiv)

Recent studies have shown remarkable progress in GANs based on implicit neural representation (INR) - an MLP that produces an RGB value given its (x, y) coordinate. They represent an image as a continuous version of the underlying 2D signal instead of a 2D array of pixels, which opens new horizons for GAN applications (e.g., zero-shot super-resolution, image outpainting). However, training existing approaches require a heavy computational cost proportional to the image resolution, since they compute an MLP operation for every (x, y) coordinate. To alleviate this issue, we propose a multi-stage patch-based training, a novel and scalable approach that can train INR-based GANs with a flexible computational cost regardless of the image resolution. Specifically, our method allows to generate and discriminate by patch to learn the local details of the image and learn global structural information by a novel reconstruction loss to enable efficient GAN training. We conduct experiments on several benchmark datasets to demonstrate that our approach enhances baseline models in GPU memory while maintaining FIDs at a reasonable level.