This repo implements Denoising Diffusion Probabilistic Models (DDPM) in Pytorch

Collection of tutorials on diffusion models, step-by-step implementation guide, scripts for generating images with AI, prompt engineering guide, and resources for further learning.

MRI Super-Resolution with Deep Learning: A Comprehensive Survey

扩散模型相关综述论文与学习资料的系统整理，涵盖基础理论、算法演化与最新研究方向。#扩散模型 #稳定扩散 #潜在扩散模型 #去噪扩散模型 #生成模型 #深度学习 #机器学习 #计算机视觉 #神经网络 #科研 #综述 #入门教程 #医学人工智能 #医疗AI #医学影像 #视网膜影像 #糖尿病视网膜病变 #图像生成 #PyTorch项目 #开源项目 #GitHub项目 #人工智能学习 #AI算法 #研究论文 #深度生成模型 #AI教育 #医疗研究 #机器学习综述

The Land-Diffuser is a novel application of the Denoising Diffusion Probabilistic Model (DDPM) in the realm of 3D Talking Head generation from raw audio inputs.

It's a simple implementation of latent diffusion models.

[ICASSP 2025]RAPID: Recognition of Any-Possible DrIver Distraction via Multi-view Pose Generation Models

[TCSVT 2025] HDiff-HIR: Hierarchically Conditional Diffusion Model for Hyperspectral Image Reconstruction

A probabilistic approach to wildfire spread prediction using a denoising diffusion model

Medical Image Synthesis project (MedSyn). In-depth evaluation of the efffects of different synthesis models (i.e., CFG ccDDPM) for medical image synthesis for class balancing on image datasets (i.e., PathMNIST).

Source code for the ICLR2026 paper titled "SPREAD: Sampling-based Pareto front Refinement via Efficient Adaptive Diffusion"

A PyTorch from-scratch implementation of Denoising Diffusion Probabilistic Models (DDPM) and Denoising Diffusion Implicit Models (DDIM) sampling to generate 16x16 pixel art sprites.

🖼️ Build and explore diffusion models for generating 16x16 pixel art sprites with a clear, from-scratch PyTorch implementation of DDPM and DDIM.

This project develops a robust medical image classification system working on latent representations of brain MRIs and leveraging Denoising Diffusion Probabilistic Models.

PyTorch implementation of Denoising Diffusion Probabilistic Models (Ho et al., 2020). Includes forward/backward diffusion process, U-Net noise predictor, and a training callback for visualising the denoising chain.

developed an adaptive X-ray denoising framework that intelligently routes each pixel through either a fast direct-mapping path (Enhanced NAFNet) or a high-fidelity diffusion path, ensuring high-quality, structurally accurate, and computationally efficient enhancement of noisy radiographs.

Diffusion models from scratch and experiments

Implementing VAEs (Variational Auto Encoders) and DDPMs (Denoising Diffusion Probabilistic Models) from-scratch, highlighting their tradeoffs and comparison through rigorous comparison

This repository is an implementation of the denoising diffusion probabilistic model described in the following paper: https://arxiv.org/abs/2006.11239

Course work from UCLA's ECE239 - Special Topics in Signals and Systems