A latent text-to-image diffusion model

The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.

Official Pytorch Implementation for "MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation" presenting "MultiDiffusion" (ICML 2023)

Official implementation for "Blended Latent Diffusion" [SIGGRAPH 2023]

A linear estimator on top of clip to predict the aesthetic quality of pictures

Referring Expression Datasets API

Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis

ADE20K Dataset

Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis

Code of the paper "FBSDiff: Plug-and-Play Frequency Band Substitution of Diffusion Features for Highly Controllable Text-Driven Image Translation" accepted by ACMMM 2024

Runway Inpainting based on Stable Diffusion

Uncovering the Text Embedding in Text-to-Image Diffusion Models