VideoFusion: Decomposed Diffusion Models for High-Quality Video Generation
This work addresses the problem of generating high-quality videos for applications in media and AI, representing an incremental improvement by adapting diffusion models with a novel decomposition to leverage temporal correlations.
The paper tackled the challenge of applying diffusion models to video generation by introducing a decomposed diffusion process that separates per-frame noise into shared base noise and time-varying residual noise, resulting in VideoFusion surpassing GAN-based and diffusion-based alternatives in high-quality video generation.
A diffusion probabilistic model (DPM), which constructs a forward diffusion process by gradually adding noise to data points and learns the reverse denoising process to generate new samples, has been shown to handle complex data distribution. Despite its recent success in image synthesis, applying DPMs to video generation is still challenging due to high-dimensional data spaces. Previous methods usually adopt a standard diffusion process, where frames in the same video clip are destroyed with independent noises, ignoring the content redundancy and temporal correlation. This work presents a decomposed diffusion process via resolving the per-frame noise into a base noise that is shared among all frames and a residual noise that varies along the time axis. The denoising pipeline employs two jointly-learned networks to match the noise decomposition accordingly. Experiments on various datasets confirm that our approach, termed as VideoFusion, surpasses both GAN-based and diffusion-based alternatives in high-quality video generation. We further show that our decomposed formulation can benefit from pre-trained image diffusion models and well-support text-conditioned video creation.