# InfinityGAN: Towards Infinite-Pixel Image Synthesis

Source: [InfinityGAN: Towards Infinite-Pixel Image Synthesis](<https://www.sota2.com/research/paper/68fa6c011e209131cea319bd>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-22T13:12:38.000Z

Authors: Chieh Hubert Lin, Hsin-Ying Lee, Yen-Chi Cheng, Sergey Tulyakov, Ming-Hsuan Yang

Published: 2021-04-08T17:59:30.000Z

## Abstract

We present a novel framework, InfinityGAN, for arbitrary-sized image generation. The task is associated with several key challenges. First, scaling existing models to an arbitrarily large image size is resource-constrained, in terms of both computation and availability of large-field-of-view training data. InfinityGAN trains and infers in a seamless patch-by-patch manner with low computational resources. Second, large images should be locally and globally consistent, avoid repetitive patterns, and look realistic. To address these, InfinityGAN disentangles global appearances, local structures, and textures. With this formulation, we can generate images with spatial size and level of details not attainable before. Experimental evaluation validates that InfinityGAN generates images with superior realism compared to baselines and features parallelizable inference. Finally, we show several applications unlocked by our approach, such as spatial style fusion, multi-modal outpainting, and image inbetweening. All applications can be operated with arbitrary input and output sizes. Please find the full version of the paper at https://openreview.net/forum?id=ufGMqIM0a4b .

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Outpainting (Right 50%) on Places2](<https://www.sota2.com/research/sota/outpainting-right-50-percent-on-places2.md>): Outpainting (Right 50%); Places2; FID: 10.6.

- [Image Uncropping on Places2 scenery categories 10k (val)](<https://www.sota2.com/research/sota/image-uncropping-on-places2-scenery-categories-10k-val.md>): Image Uncropping; Places2 scenery categories 10k (val); FID: 15.7.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2104.03963>)

[PDF](<https://arxiv.org/pdf/2104.03963.pdf>)