Scalable High-Resolution Pixel-Space Image Synthesis with Hourglass Diffusion Transformers
This addresses the challenge of scalable high-resolution image generation for AI and computer vision applications, representing a novel method rather than an incremental improvement.
The paper tackles the problem of high-resolution image synthesis by introducing the Hourglass Diffusion Transformer (HDiT), which achieves linear scaling with pixel count and sets a new state-of-the-art for diffusion models on FFHQ-1024^2.
We present the Hourglass Diffusion Transformer (HDiT), an image generative model that exhibits linear scaling with pixel count, supporting training at high-resolution (e.g. $1024 \times 1024$) directly in pixel-space. Building on the Transformer architecture, which is known to scale to billions of parameters, it bridges the gap between the efficiency of convolutional U-Nets and the scalability of Transformers. HDiT trains successfully without typical high-resolution training techniques such as multiscale architectures, latent autoencoders or self-conditioning. We demonstrate that HDiT performs competitively with existing models on ImageNet $256^2$, and sets a new state-of-the-art for diffusion models on FFHQ-$1024^2$.