Denoising Diffusion Probabilistic Models

Jonathan Ho, Ajay Jain, Pieter Abbeel
6/19/2020
cs.LGstat.ML

Official code: hojonathanho/diffusion (5,343 stars, last commit 2023-08-29).

Abstract

We present high quality image synthesis results using diffusion probabilistic models, a class of latent variable models inspired by considerations from nonequilibrium thermodynamics. Our best results are obtained by training on a weighted variational bound designed according to a novel connection between diffusion probabilistic models and denoising score matching with Langevin dynamics, and our models naturally admit a progressive lossy decompression scheme that can be interpreted as a generalization of autoregressive decoding. On the unconditional CIFAR10 dataset, we obtain an Inception score of 9.46 and a state-of-the-art FID score of 3.17. On 256x256 LSUN, we obtain sample quality similar to ProgressiveGAN. Our implementation is available at https://github.com/hojonathanho/diffusion

View on arXivView PDF

Code Implementations(1)

hojonathanho/diffusionOfficial100%

Denoising Diffusion Probabilistic Models

5,343494PythonJun 19, 20203 years ago

Similar Papers

DiffuseVAE: Efficient, Controllable and High-Fidelity Generation from Low-Dimensional Latents

88% match1 repo3832022

Denoising Diffusion Implicit Models

87% match1 repo1.9k2020

VideoFusion: Decomposed Diffusion Models for High-Quality Video Generation

87% match2023

Revisiting Non-Autoregressive Transformers for Efficient Image Synthesis

87% match1 repo472024

Look ATME: The Discriminator Mean Entropy Needs Attention

86% match1 repo292023

Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes

86% match1 repo1862021

Cite this paper

@article{ho2020denoising,
  title  = {Denoising Diffusion Probabilistic Models},
  author = {Jonathan Ho and Ajay Jain and Pieter Abbeel},
  year   = {2020},
  eprint = {2006.11239},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2006.11239}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2006.11239.svg)](https://codeofpaper.com/paper/2006.11239)

Discussion