Sometimes you are a member of a group of people who want to create images with generative AI that serve some practical purpose, such as illustrating the ideas in a slide presentation or a website.
So the goal is not to create amazing “AI art” but to avoid having to search desperately in stock image collections, hire an artist who is available 24/7, etc.
But even with the best image generators, it usually requires a good deal trial and error to arrive at images that illustrate your ideas clearly and in an appealing style. The image generator has at best limited understanding of your intentions with each image and with the presentation as a whole, so you have to keep nudging it in the right direction.
If you’ve tried ad hoc methods like having group members paste generated images into shared documents, you will have noticed that the process tends to be unstructured, inefficient, and ineffective.
Enter COLL-E, which was designed specifically to support this type of collaboration within the more general Groupsheets platform.
COLL-E currently makes use of the latest available versions of OpenAI's DALL-E (for image generation) and ChatGPT (for help with prompt formulation). Other sources of image and text generation may be included in the future.
The screenshots below show how COLL-E offers unique forms of support.
In the example, two colleagues, Anna and Peter, are creating a slide presentation with advice for students who are about to attend a professional conference for the first time.
(They can work in the same or different locations, at the same or different times. In these examples, they request just 2 images at a time.)