Abstract:Denoising Diffusion Probabilistic Models (DDPM) have recently gained significant attention. DDPMs compose a Markovian process that begins in the data domain and gradually adds noise until reaching pure white noise. DDPMs generate high-quality samples from complex data distributions by defining an inverse process and training a deep neural network to learn this mapping. However, these models are inefficient because they require many diffusion steps to produce aesthetically pleasing samples. Additionally, unlike generative adversarial networks (GANs), the latent space of diffusion models is less interpretable. In this work, we propose to generalize the denoising diffusion process into an Upsampling Diffusion Probabilistic Model (UDPM). In the forward process, we reduce the latent variable dimension through downsampling, followed by the traditional noise perturbation. As a result, the reverse process gradually denoises and upsamples the latent variable to produce a sample from the data distribution. We formalize the Markovian diffusion processes of UDPM and demonstrate its generation capabilities on the popular FFHQ, AFHQv2, and CIFAR10 datasets. UDPM generates images with as few as three network evaluations, whose overall computational cost is less than a single DDPM or EDM step, while achieving an FID score of 6.86. This surpasses current state-of-the-art efficient diffusion models that use a single denoising step for sampling. Additionally, UDPM offers an interpretable and interpolable latent space, which gives it an advantage over traditional DDPMs. Our code is available online: \url{<a class="link-external link-https" href="https://github.com/shadyabh/UDPM/" rel="external noopener nofollow">this https URL</a>}

Diffusion Models Without Attention

Denoising Diffusion Step-aware Models

Attention-Driven Training-Free Efficiency Enhancement of Diffusion Models

Scalable Diffusion Models with State Space Backbone

FAM Diffusion: Frequency and Attention Modulation for High-Resolution Image Generation with Stable Diffusion

Directly Denoising Diffusion Models

Enhancing Diffusion Models for High-Quality Image Generation

HiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Models

Diff-2-in-1: Bridging Generation and Dense Perception with Diffusion Models

The Uncanny Valley: A Comprehensive Analysis of Diffusion Models

Diffusion Models Need Visual Priors for Image Generation

simple diffusion: End-to-end diffusion for high resolution images

PartDiff: Image Super-resolution with Partial Diffusion Models

Stimulating Diffusion Model for Image Denoising via Adaptive Embedding and Ensembling

The Missing U for Efficient Diffusion Models

High-Resolution Image Synthesis with Latent Diffusion Models

Stimulating the Diffusion Model for Image Denoising Via Adaptive Embedding and Ensembling

Fixed Point Diffusion Models

ACDMSR: Accelerated Conditional Diffusion Models for Single Image Super-Resolution

UDPM: Upsampling Diffusion Probabilistic Models