D

Ddpm Celebahq 256

Developed by google
High-quality image generation model based on diffusion probabilistic models, excelling in unconditional image generation tasks
Downloads 21.82k
Release Time : 7/19/2022

Model Overview

DDPM is a latent variable model inspired by non-equilibrium thermodynamics, achieving high-quality image synthesis through a gradual denoising process and supporting progressive lossy decompression

Model Features

High-quality image generation
Achieves state-of-the-art Inception and FID scores on datasets like CIFAR10 and LSUN
Multi-scheduler support
Supports various noise schedulers including DDPM, DDIM, and PNDM, allowing flexible choices between generation quality and speed
Progressive generation
Naturally supports progressive lossy decompression schemes, which can be viewed as a generalization of autoregressive decoding

Model Capabilities

Unconditional image generation
High-quality face synthesis
Progressive image generation

Use Cases

Image generation
Celebrity face generation
Generates high-quality celebrity face images based on the CelebA-HQ dataset
Produces high-quality face images at 256x256 resolution
Art creation
Used for digital art creation and concept design
Can generate diverse visual content
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase