Contrastive Sequential-Diffusion Learning: Non-linear and Multi-Scene Instructional Video Synthesis

TL;DR

This paper introduces a contrastive sequential diffusion method for multi-scene instructional video synthesis, ensuring visual consistency across non-linear scene sequences based on scene descriptions.

Contribution

It presents a novel contrastive learning approach that selects relevant previous scenes to guide the generation of subsequent scenes, improving multi-scene video coherence.

Findings

Enhanced scene consistency in generated videos

Outperforms previous methods in real-world action data

Effective handling of non-linear scene sequences

Abstract

Generated video scenes for action-centric sequence descriptions, such as recipe instructions and do-it-yourself projects, often include non-linear patterns, where the next video may need to be visually consistent not with the immediately preceding video but with earlier ones. Current multi-scene video synthesis approaches fail to meet these consistency requirements. To address this, we propose a contrastive sequential video diffusion method that selects the most suitable previously generated scene to guide and condition the denoising process of the next scene. The result is a multi-scene video that is grounded in the scene descriptions and coherent w.r.t. the scenes that require visual consistency. Experiments with action-centered data from the real world demonstrate the practicality and improved consistency of our model compared to previous work.