Hierarchical Spatio-temporal Decoupling for Text-to-Video Generation

TL;DR

HiGen introduces a hierarchical decoupling approach for text-to-video generation, separating spatial and temporal factors at structure and content levels to improve realism, diversity, and stability of generated videos.

Contribution

The paper proposes a novel hierarchical decoupling framework that separates spatial and temporal reasoning, enhancing the quality and stability of text-to-video synthesis.

Findings

Outperforms state-of-the-art T2V methods in quality and stability

Effectively reduces complexity of video generation task

Generates semantically accurate and temporally stable videos

Abstract

Despite diffusion models having shown powerful abilities to generate photorealistic images, generating videos that are realistic and diverse still remains in its infancy. One of the key reasons is that current methods intertwine spatial content and temporal dynamics together, leading to a notably increased complexity of text-to-video generation (T2V). In this work, we propose HiGen, a diffusion model-based method that improves performance by decoupling the spatial and temporal factors of videos from two perspectives, i.e., structure level and content level. At the structure level, we decompose the T2V task into two steps, including spatial reasoning and temporal reasoning, using a unified denoiser. Specifically, we generate spatially coherent priors using text during spatial reasoning and then generate temporally coherent motions from these priors during temporal reasoning. At the content level, we extract two subtle cues from the content of the input video that can express motion and appearance changes, respectively. These two cues then guide the model's training for generating videos, enabling flexible content variations and enhancing temporal stability. Through the decoupled paradigm, HiGen can effectively reduce the complexity of this task and generate realistic videos with semantics accuracy and motion stability. Extensive experiments demonstrate the superior performance of HiGen over the state-of-the-art T2V methods.