CVApr 8, 2021

InfinityGAN: Towards Infinite-Pixel Image Synthesis

arXiv:2104.03963v485 citations
Originality Highly original
AI Analysis

This addresses the challenge of infinite-pixel image synthesis for computer vision and graphics applications, offering a novel method for scalable generation.

The paper tackles the problem of generating arbitrarily large images by proposing InfinityGAN, a framework that trains and infers patch-by-patch with low computational resources, resulting in images with superior realism compared to baselines and enabling applications like spatial style fusion and multi-modal outpainting.

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 .

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes