StyleCineGAN: Landscape Cinemagraph Generation using a Pre-trained StyleGAN

TL;DR

StyleCineGAN introduces a novel approach for automatic landscape cinemagraph generation from still images by leveraging deep feature warping in a pre-trained StyleGAN, producing high-quality, looping animations.

Contribution

It utilizes deep feature space warping in StyleGAN for high-resolution cinemagraph synthesis, a novel technique compared to prior latent space methods.

Findings

Produces high-resolution, plausible looping cinemagraphs

Outperforms state-of-the-art methods in user studies

Demonstrates effectiveness of deep feature warping in cinemagraph generation

Abstract

We propose a method that can generate cinemagraphs automatically from a still landscape image using a pre-trained StyleGAN. Inspired by the success of recent unconditional video generation, we leverage a powerful pre-trained image generator to synthesize high-quality cinemagraphs. Unlike previous approaches that mainly utilize the latent space of a pre-trained StyleGAN, our approach utilizes its deep feature space for both GAN inversion and cinemagraph generation. Specifically, we propose multi-scale deep feature warping (MSDFW), which warps the intermediate features of a pre-trained StyleGAN at different resolutions. By using MSDFW, the generated cinemagraphs are of high resolution and exhibit plausible looping animation. We demonstrate the superiority of our method through user studies and quantitative comparisons with state-of-the-art cinemagraph generation methods and a video generation method that uses a pre-trained StyleGAN.