Blowing in the Wind: CycleNet for Human Cinemagraphs from Still Images
This addresses the problem of tedious manual creation of cinemagraphs for artists and media creators, though it is incremental as it focuses on a specific domain (dressed humans under wind).
The paper tackles automatic generation of human cinemagraphs (looping videos with subtle motion) from single RGB images, specifically for dressed humans under wind, by proposing a cyclic neural network that learns from synthetic data and generalizes to real data, producing compelling results.
Cinemagraphs are short looping videos created by adding subtle motions to a static image. This kind of media is popular and engaging. However, automatic generation of cinemagraphs is an underexplored area and current solutions require tedious low-level manual authoring by artists. In this paper, we present an automatic method that allows generating human cinemagraphs from single RGB images. We investigate the problem in the context of dressed humans under the wind. At the core of our method is a novel cyclic neural network that produces looping cinemagraphs for the target loop duration. To circumvent the problem of collecting real data, we demonstrate that it is possible, by working in the image normal space, to learn garment motion dynamics on synthetic data and generalize to real data. We evaluate our method on both synthetic and real data and demonstrate that it is possible to create compelling and plausible cinemagraphs from single RGB images.