A Good Image Generator Is What You Need for High-Resolution Video Synthesis

TL;DR

This paper introduces a novel framework that uses pre-trained image generators to efficiently produce high-resolution videos by discovering trajectories in the latent space, enabling content and motion manipulation and cross-domain synthesis.

Contribution

It presents a new method for high-resolution video synthesis leveraging image generators, including a motion generator for disentangled content and motion, and introduces cross-domain video synthesis.

Findings

Outperforms existing video generation methods in quality and efficiency

Enables content and motion manipulation in generated videos

Supports cross-domain video synthesis with disjoint datasets

Abstract

Image and video synthesis are closely related areas aiming at generating content from noise. While rapid progress has been demonstrated in improving image-based models to handle large resolutions, high-quality renderings, and wide variations in image content, achieving comparable video generation results remains problematic. We present a framework that leverages contemporary image generators to render high-resolution videos. We frame the video synthesis problem as discovering a trajectory in the latent space of a pre-trained and fixed image generator. Not only does such a framework render high-resolution videos, but it also is an order of magnitude more computationally efficient. We introduce a motion generator that discovers the desired trajectory, in which content and motion are disentangled. With such a representation, our framework allows for a broad range of applications, including content and motion manipulation. Furthermore, we introduce a new task, which we call cross-domain video synthesis, in which the image and motion generators are trained on disjoint datasets belonging to different domains. This allows for generating moving objects for which the desired video data is not available. Extensive experiments on various datasets demonstrate the advantages of our methods over existing video generation techniques. Code will be released at https://github.com/snap-research/MoCoGAN-HD.