Tell Me What Happened: Unifying Text-guided Video Completion via Multimodal Masked Video Generation

TL;DR

This paper introduces a unified model, MMVG, for text-guided video completion tasks including prediction, rewind, and infilling, by leveraging masked video generation and visual token discretization.

Contribution

The paper proposes a novel multimodal masked video generation approach that unifies various video completion tasks under a single framework guided by natural language.

Findings

Effective in diverse scenarios including egocentric, animation, and gaming videos.

Generates high-quality visual appearances aligned with text instructions.

Single model handles prediction, rewind, and infilling tasks seamlessly.

Abstract

Generating a video given the first several static frames is challenging as it anticipates reasonable future frames with temporal coherence. Besides video prediction, the ability to rewind from the last frame or infilling between the head and tail is also crucial, but they have rarely been explored for video completion. Since there could be different outcomes from the hints of just a few frames, a system that can follow natural language to perform video completion may significantly improve controllability. Inspired by this, we introduce a novel task, text-guided video completion (TVC), which requests the model to generate a video from partial frames guided by an instruction. We then propose Multimodal Masked Video Generation (MMVG) to address this TVC task. During training, MMVG discretizes the video frames into visual tokens and masks most of them to perform video completion from any time point. At inference time, a single MMVG model can address all 3 cases of TVC, including video prediction, rewind, and infilling, by applying corresponding masking conditions. We evaluate MMVG in various video scenarios, including egocentric, animation, and gaming. Extensive experimental results indicate that MMVG is effective in generating high-quality visual appearances with text guidance for TVC.