Stable Diffusion

Image-to-Text AI Models: CLIP, BLIP, WD 1.4 (aka WD14), and GPT-4V (Vision)

Automated tagging, labeling, or describing of images is a crucial task in many applications, particularly in the preparation of datasets for machine learning. This is where image-to-text models come to the rescue. Among the leading…

Upscaling Images in Automatic1111: Tiled VAE and Multidiffusion Upscaler

Creating high-resolution images in Automatic1111 has never been easier, thanks to the Tiled Variational Autoencoder (VAE) and Multidiffusion Upscaler. These powerful tools enable users to generate impressive 4k+ images without visible seams or complicated steps.…

AI Can Read Your Mind: Stable Diffusion and the Future of Brain-Computer Interfaces

Presented images (red box, top row) and images reconstructed from fMRI signals (gray box, bottom row) for one subject.

Today, we will be discussing a groundbreaking paper that explores how researchers used stable diffusion, a type of generative AI, to reconstruct images based on human brain activity. This research not only has significant implications…