GenTron: Diffusion Transformers for Image and Video Generation

TL;DR

GenTron introduces Transformer-based diffusion models for image and video generation, achieving superior visual quality and text alignment, and extending the application of diffusion Transformers to visual content.

Contribution

This work pioneers the use of Transformer-based diffusion models for image and video generation, scaling models significantly and introducing novel motion-free guidance for videos.

Findings

GenTron outperforms SDXL in visual quality and text alignment.

Scaling from 900M to 3B parameters improves visual results.

GenTron excels in compositional generation benchmarks.

Abstract

In this study, we explore Transformer-based diffusion models for image and video generation. Despite the dominance of Transformer architectures in various fields due to their flexibility and scalability, the visual generative domain primarily utilizes CNN-based U-Net architectures, particularly in diffusion-based models. We introduce GenTron, a family of Generative models employing Transformer-based diffusion, to address this gap. Our initial step was to adapt Diffusion Transformers (DiTs) from class to text conditioning, a process involving thorough empirical exploration of the conditioning mechanism. We then scale GenTron from approximately 900M to over 3B parameters, observing significant improvements in visual quality. Furthermore, we extend GenTron to text-to-video generation, incorporating novel motion-free guidance to enhance video quality. In human evaluations against SDXL, GenTron achieves a 51.1% win rate in visual quality (with a 19.8% draw rate), and a 42.3% win rate in text alignment (with a 42.9% draw rate). GenTron also excels in the T2I-CompBench, underscoring its strengths in compositional generation. We believe this work will provide meaningful insights and serve as a valuable reference for future research.