AI image generation is a technology that has been hotly discussed in the art and Deep Learning (DL) field. You must have heard of the AI Art Generator such as Dall-E 2 or NovelAI, a DL model that generates realistic-looking images from a given text sequence. 
To explore this technology deeper, we need to introduce a new class in the generative model called 'diffusion', first proposed by Sohl-Dickstein et al. (2015), which aimed to generate images from noise using a backward denoising process.

Thus, our target is to find the noise (movement) added to the image and reverse this process. The diffusion model is mainly composed of two processes Forward Noising and Backward Denoising; this could be regarded as continuously adding noise into the image than reversing this process.

For the detail introduction, welcome visit my medium [link]

data was collected from kaggle (another anime data)
(https://www.kaggle.com/datasets/scribbless/another-anime-face-dataset)

Denoising Diffusion Probabilistic Models Denoising Diffusion Implicit Models Keras example - Denoising Diffusion Implicit Models Lil' Log - What are Diffusion Models? Vedant Jumle - Image generation with diffusion models using Keras and TensorFlow

1. Denoising Diffusion Probabilistic Models [link]
2. Denoising Diffusion Implicit Models [link]
3. Keras example - Denoising Diffusion Implicit Models [link]
4. Lil' Log - What are Diffusion Models? [link]
5. Vedant Jumle - Image generation with diffusion models using Keras and TensorFlow [link]