Uni-Classifier: Leveraging Video Diffusion Priors for Universal Guidance Classifier

TL;DR

Uni-Classifier (Uni-C) is a versatile plug-and-play module that uses video diffusion priors to improve the quality and alignment of generative model outputs in complex AI workflows, enhancing downstream task performance.

Contribution

This paper introduces Uni-Classifier, a novel method leveraging video diffusion priors to guide and improve generative model outputs in a plug-and-play manner.

Findings

Consistently improves generation quality across video and 3D tasks.

Effective in both workflow-based and standalone applications.

Demonstrates strong generalization across different generative models.

Abstract

In practical AI workflows, complex tasks often involve chaining multiple generative models, such as using a video or 3D generation model after a 2D image generator. However, distributional mismatches between the output of upstream models and the expected input of downstream models frequently degrade overall generation quality. To address this issue, we propose Uni-Classifier (Uni-C), a simple yet effective plug-and-play module that leverages video diffusion priors to guide the denoising process of preceding models, thereby aligning their outputs with downstream requirements. Uni-C can also be applied independently to enhance the output quality of individual generative models. Extensive experiments across video and 3D generation tasks demonstrate that Uni-C consistently improves generation quality in both workflow-based and standalone settings, highlighting its versatility and strong generalization capability.