Tensorflow Diffusion is an experimental project that is built to showcase the new Stable Diffusion Model in TensorFlow that is built with Keras CV it showcases the ability of the model to generate high-resolution images. It uses Latent Diffusion and Efficient Sub-Pixel CNN for generating High-resolution images
It uses a Text encoder to encode the image based on the requirements uses noise to pass on to the Diffusion model and uses a Decoder to convert it into a high-resolution image.