LearnAI x WebinarsGenerative imaging in Midjourney and Dall-E
Generative imaging with Midjourney

TheFactor.e was one of the speakers during our online event: The Webinarverse. 

Martijn Loonstra is a Data Consultant at TheFactor.e. TheFactor.e has over 25 years of experience in the digital field, working with large partners like Sitecore, Umbraco, Bloomreach, Azure, and AWS. They are based in Groningen with more than 85 colleagues and focus on sectors like education, employment, and health

Midjourney and Dall-E

In the webinar, he focused on Midjourney and Dall-E, two state-of-the-art models for generative imaging AI. Midjourney is a diffusion model, while Dall-E is a text-based deep neural network. There are many free alternatives, such as Leonardo.ai.

Prompting and parameters

Prompts are crucial for directing the AI in creating the desired images. 

Midjourney creates images based on visual data using a diffusion process, resulting in beautiful, artsy images. Dall-E, on the other hand, understands text prompts better, producing more accurate interpretations of descriptions. In the webinar he demonstrates this with examples, highlighting the strengths and weaknesses of each model.

Using AI for realistic and artistic images

You can use various prompts to generate realistic images that look like they were taken in real life. This includes leveraging parameters like lighting, angles, and color palettes to enhance the realism of your AI-generated images. Additionally, you can create artistic images by specifying styles, lighting effects, and other visual elements.

Future developments and conclusion

Generative imaging AI offers a large range of possibilities for creating engaging and visually stunning content. Whether you're using Midjourney for its artistic flair or Dall-E for its textual understanding, the key is to experiment with prompts and parameters to achieve the desired results.

Watch full webinar on demand
View webinar

Related articles