Mind the Time: Temporally-Controlled Multi-Event Video Generation

TL;DR

MinT is a novel multi-event video generation model that enables precise temporal control over events within videos, producing coherent sequences by binding events to specific time periods and using a new time-aware encoding.

Contribution

We introduce MinT, the first model to provide temporal control over multiple events in video generation, utilizing a novel time-based positional encoding called ReRoPE.

Findings

MinT outperforms existing models in generating temporally controlled videos.

The ReRoPE encoding effectively guides cross-attention for temporal coherence.

Experiments show improved event ordering and timing accuracy.

Abstract

Real-world videos consist of sequences of events. Generating such sequences with precise temporal control is infeasible with existing video generators that rely on a single paragraph of text as input. When tasked with generating multiple events described using a single prompt, such methods often ignore some of the events or fail to arrange them in the correct order. To address this limitation, we present MinT, a multi-event video generator with temporal control. Our key insight is to bind each event to a specific period in the generated video, which allows the model to focus on one event at a time. To enable time-aware interactions between event captions and video tokens, we design a time-based positional encoding method, dubbed ReRoPE. This encoding helps to guide the cross-attention operation. By fine-tuning a pre-trained video diffusion transformer on temporally grounded data, our approach produces coherent videos with smoothly connected events. For the first time in the literature, our model offers control over the timing of events in generated videos. Extensive experiments demonstrate that MinT outperforms existing commercial and open-source models by a large margin.