SparseCtrl: Adding Sparse Controls to Text-to-Video Diffusion Models

TL;DR

SparseCtrl introduces a method for controlling text-to-video generation using sparse structural signals, enhancing flexibility and practicality without retraining the underlying models.

Contribution

It enables structure control with minimal inputs by adding a condition encoder to pre-trained models, compatible with various modalities.

Findings

Effective control with sparse signals demonstrated

Compatible with multiple modalities like sketches and depth maps

Generalizes across different text-to-video models

Abstract

The development of text-to-video (T2V), i.e., generating videos with a given text prompt, has been significantly advanced in recent years. However, relying solely on text prompts often results in ambiguous frame composition due to spatial uncertainty. The research community thus leverages the dense structure signals, e.g., per-frame depth/edge sequences, to enhance controllability, whose collection accordingly increases the burden of inference. In this work, we present SparseCtrl to enable flexible structure control with temporally sparse signals, requiring only one or a few inputs, as shown in Figure 1. It incorporates an additional condition encoder to process these sparse signals while leaving the pre-trained T2V model untouched. The proposed approach is compatible with various modalities, including sketches, depth maps, and RGB images, providing more practical control for video generation and promoting applications such as storyboarding, depth rendering, keyframe animation, and interpolation. Extensive experiments demonstrate the generalization of SparseCtrl on both original and personalized T2V generators. Codes and models will be publicly available at https://guoyww.github.io/projects/SparseCtrl .