Diffusion Models for Video Prediction and Infilling

TL;DR

This paper introduces RaMViD, a diffusion model extension for videos that uses 3D convolutions and a novel conditioning technique, enabling state-of-the-art video prediction, infilling, and upsampling.

Contribution

The paper presents RaMViD, a unified diffusion model for videos that supports multiple tasks with a simple conditioning scheme, trained in both conditional and unconditional modes.

Findings

Achieves state-of-the-art results on video prediction benchmarks.

Effectively performs video infilling and upsampling.

Supports high-resolution video generation.

Abstract

Predicting and anticipating future outcomes or reasoning about missing information in a sequence are critical skills for agents to be able to make intelligent decisions. This requires strong, temporally coherent generative capabilities. Diffusion models have shown remarkable success in several generative tasks, but have not been extensively explored in the video domain. We present Random-Mask Video Diffusion (RaMViD), which extends image diffusion models to videos using 3D convolutions, and introduces a new conditioning technique during training. By varying the mask we condition on, the model is able to perform video prediction, infilling, and upsampling. Due to our simple conditioning scheme, we can utilize the same architecture as used for unconditional training, which allows us to train the model in a conditional and unconditional fashion at the same time. We evaluate RaMViD on two benchmark datasets for video prediction, on which we achieve state-of-the-art results, and one for video generation. High-resolution videos are provided at https://sites.google.com/view/video-diffusion-prediction.