CVAIGRMar 21, 2024

StyleCineGAN: Landscape Cinemagraph Generation using a Pre-trained StyleGAN

arXiv:2403.14186v17 citationsh-index: 7CVPR
Originality Incremental advance
AI Analysis

This work addresses the problem of automated cinemagraph creation for landscape images, offering a domain-specific incremental improvement over existing techniques.

The paper tackles generating cinemagraphs from still landscape images by using a pre-trained StyleGAN, achieving high-resolution and plausible looping animations through a novel multi-scale deep feature warping method, with results validated by user studies and quantitative comparisons against state-of-the-art methods.

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.

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