TL;DR
This paper presents a novel method for generating infinite, high-resolution, and diverse images by aligning latent and image spaces, enabling seamless connection of unrelated scenes into large panoramas.
Contribution
The authors introduce a perfectly equivariant generator with synchronized interpolation in latent and image spaces, and modify AdaIN for this setup, enabling new image generation capabilities.
Findings
Outperforms baselines by at least 4 times in quality and diversity
Enables connecting unrelated scenes into large panoramas
Successfully generates infinite high-resolution images
Abstract
In this work, we develop a method to generate infinite high-resolution images with diverse and complex content. It is based on a perfectly equivariant generator with synchronous interpolations in the image and latent spaces. Latent codes, when sampled, are positioned on the coordinate grid, and each pixel is computed from an interpolation of the nearby style codes. We modify the AdaIN mechanism to work in such a setup and train the generator in an adversarial setting to produce images positioned between any two latent vectors. At test time, this allows for generating complex and diverse infinite images and connecting any two unrelated scenes into a single arbitrarily large panorama. Apart from that, we introduce LHQ: a new dataset of \lhqsize high-resolution nature landscapes. We test the approach on LHQ, LSUN Tower and LSUN Bridge and outperform the baselines by at least 4 times in…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
MethodsAdaptive Instance Normalization · Aligning Latent and Image Spaces
