GenTron: Diffusion Transformers for Image and Video Generation
This work addresses the problem of improving generative models for researchers and practitioners by introducing a novel Transformer-based approach, though it is incremental as it adapts existing diffusion methods to new architectures.
The study tackled the gap in using Transformer-based architectures for image and video generation by introducing GenTron, a family of diffusion models, which achieved a 51.1% win rate in visual quality and 42.3% win rate in text alignment against SDXL in human evaluations.
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.