Reader, I confess….I have started to get carried away with the various different ways AI can enhance my work- whether through creating learning sessions, summarising reports, taking meeting notes…the possibilities are infinite. This week I disappeared down another appealing rabbit hole to explore the creation of AI images. Ostensibly to enhance my presentations, but mainly as I love ‘colouring in’ and finding the right image to fit my blog posts.
Here I will outline how you can harness DALL-E, which ‘can create realistic images and art from a description in natural language.’
You can access both Chat GPT 3.5 and DALL-E through Open AI and Bing. However, DALL-E is FREE through Bing- and Chat GPT is also version 4, so the most recent version.
Good prompts are key
I then looked on line to see what other suggestions there were for creating good visual prompts and then decided (doh!) to ask Chat GPT. So before you too decide to go and get creative, here are Chat GPT’s suggestions:
Creating effective prompts for generating high-quality images with DALL-E can be a creative and iterative process. DALL-E responds well to specific and well-structured prompts. Here are some tips to help you craft prompts that yield good quality images:
- Be Clear and Specific: Provide a clear and specific description of what you want in the image. The more precise your prompt, the better the chances of getting the desired output. For example, instead of saying “a cat,” specify “a Siamese cat with a red ribbon around its neck sitting on a windowsill.”
- Use Adjectives and Nouns: Combine descriptive adjectives with nouns to refine your prompt. For example, “a futuristic cityscape with towering skyscrapers” provides a more vivid description than simply saying “city.”
- Set the Context: Add context or a scenario to your prompt. Describe the setting or the scene to help DALL-E understand the context better. For instance, “a cozy cabin in the woods with a crackling fireplace” is more informative than “cabin.”
- Experiment with Size and Quantity: You can specify the size, quantity, or arrangement of objects in the image. For example, “a field of a thousand sunflowers” or “a tiny, intricate pocket watch.”
- Use Analogies and References: Draw analogies or make references to well-known things or concepts. This can help DALL-E understand your prompt. For example, “a car with the elegance of a Rolls-Royce and the speed of a Ferrari.”
- Combine Concepts: Combine two or more unrelated concepts to create something unique and imaginative. This can result in creative and surprising images. For example, “a tree made of books” or “a teapot shaped like a rocket ship.”
- Experiment with Color and Style: Specify colors or artistic styles if they are crucial to your prompt. For example, “a monochromatic painting of a red rose” or “a cubist interpretation of a city skyline in shades of blue.”
- Avoid Ambiguity: Try to eliminate ambiguity from your prompts. Be precise about what you want, so there’s no room for interpretation. If something could be interpreted in multiple ways, clarify it.
- Iterate and Experiment: Don’t be afraid to iterate and experiment with different prompts. DALL-E may produce varying results based on the wording, so try multiple prompts to see which one works best.
- Consider Open-Ended Prompts: If you want to see DALL-E’s creative abilities, you can use more open-ended prompts like, “Create an image that represents the concept of ‘freedom‘.”
- Provide Constraints: If you have specific preferences or constraints, make them clear in your prompt. For example, “Create a grayscale image of a vintage bicycle with no people in.”
If you enjoy the process of creating unique, unusual and memorable images you could have a lot of fun with DALL-E. In the meantime, I hope you enjoyed this week’s bog visual, which was generated using the prompt: “Alice in wonderland disappearing down a rabbit hole in the artistic style of Salvador Dali.”
