In the year 2022 AI driven image generation reached a level of maturity that many people would feel happy with the created images. DALL-E 2, MidJourney, Stable Diffusion are the most commonly known examples.
2023 and the following years those technologies are becoming more and more pervasive and it is important for everybody to take a closer look.
In this talk, we will look at what are some of the more prominent image generation tools. Without going into the deep levels of the underlying math, the talk will briefly touch the AI approaches behind text guided image generation.
Many of the AI models are only accessible to individuals and business through consuming a cloud-based API, but there are a few options where businesses and individuals can work in their own environments, for example "Stable Diffusion", which we will cover.
We will then take it a step further and look at some of the more advanced scenarios that are possible, for example optimizing image output for a specific style or modify a model so it generates images that look like you.
This session focuses on practical approaches to this domain, but some of the risks and trends in dealing with risks in the context of Responsible AI will be covered.
You will learn:
- About the state of the art of AI driven image generation
- Understand the typical building blocks of AI driven image generation
- Prepare yourself to start generating images