Scalable Diffusion Models with Transformers

By William S. Peebles, Saining Xie. IEEE International Conference on Computer Vision, 2022

6,663 citations recorded by the source corpus.

We explore a new class of diffusion models based on the transformer architecture. We train latent diffusion models of images, replacing the commonly-used U-Net backbone with a transformer that operates on latent patches. We analyze the scalability of our Diffusion Transformers (DiTs) through the lens of forward pass complexity as measured by Gflops. We find that DiTs with higher Gflops—through increased transformer depth/width or increased number of input tokens—consistently have lower FID. In addition to possessing good scalability properties, our largest DiT-XL/2 models outperform all prior diffusion models on the class-conditional ImageNet 512×512 and 256×256 benchmarks, achieving a state-of-the-art FID of 2.27 on the latter.

Subfields: Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging, Generative Models, Model Architectures, Signal Processing

View Paper: https://arxiv.org/abs/2212.09748

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

  1. Luc Van Gool
  2. Ming-Hsuan Yang
  3. Dacheng Tao
  4. Xiaogang Wang
  5. Jitendra Malik
  6. Trevor Darrell
  7. Andrew Zisserman
  8. Andrea Vedaldi

In Radiology, Nuclear Medicine and Imaging

  1. Nicholas Matthew Boffi
  2. Eric Vanden-Eijnden
  3. Michael S. Albergo
  4. Diederik P. Kingma
  5. Ruiqi Gao
  6. Georgios Pavlakos
  7. Ming-Hsuan Yang
  8. Jitendra Malik

In Generative Models

  1. Richard Zhang
  2. Eli Shechtman
  3. Andreas Blattmann
  4. Robin Rombach
  5. Jun-Yan Zhu
  6. Stefano Ermon
  7. Ming-Hsuan Yang
  8. David J. Fleet

In Model Architectures

  1. Ping Luo
  2. Pan Zhou
  3. Shuicheng Yan
  4. Yu Qiao
  5. Jian‐Yuan Sun
  6. Han Hu
  7. Ming-Hsuan Yang
  8. Saining Xie

In Signal Processing

  1. Stefano Ermon
  2. Robin Rombach
  3. David J. Fleet
  4. Richard Zhang
  5. Ming-Hsuan Yang
  6. Andreas Blattmann
  7. Patrick Esser
  8. Eli Shechtman

Citations in the corpus