Studio Ghibli-style images created using ChatGPT’s native image generator have dominated the headlines for the past few days, but social media users seem to have finally come to terms with the fact that there is much more that can be done with OpenAI’s new tool. After turning their images into the iconic Japanese art style, users are now rushing to create a realistic action figure of themselves. If you don’t want to miss out on the trend, here’s a step-by-step guide on how to get started.
How to generate realistic action figure like images of yourself using ChatGPT?
1) Open the ChatGPT app or website
2) Start a new conversation and make sure that the AI model is selected as GPT-4o
3) Upload a picture of yourself before writing the following text prompt:
“Using the photo of me that I will upload, create a realistic action figure of myself in a blister pack, styled like a premium collectible toy. The figure should be posed standing upright. The blister pack should have a red header with the text ‘[Your Name]’ in large white letters, and below it, ‘Your designation’ in smaller white letters. Add an ‘Ages 17+’ label in the top right corner of the header. Include accessories in compartments on the right side of the figure: a notebook, a pen, a small camera, and a laptop with a ChatGPT logo on it. The background of the blister pack should be beige. Ensure the action figure retains my facial features and general appearance from the uploaded photo, with a serious expression, and render the image in high detail with photorealistic quality.”
Here’s the realistic action figure like image we generated using ChatGPT:
4) Feel free to make any changes to the image by tweaking the text prompt or asking ChatGPT for the edits after the image is generated.
How are the results with ChatGPT’s action figure-like images?
While ChatGPT managed to get most of the other required details right (including all the text elements), it failed to get my facial features even remotely right. The resulting image looks like a generic brown man with a beard wearing clothes similar to mine.
However, many other users have had better success with this style and my uploaded image may not be the best for this particular subject. In any case, readers will have to try it for themselves to see how it works.
Why is there a sudden trend for ChatGPT-generated images?
ChatGPT has been able to generate images from text prompts for some time, but the trend for its images started after OpenAI unlocked native image generation capabilities for the chatbot. What this meant was that GPT-4o (the base model that runs ChatGPT for most users) could now also generate images directly instead of relying on an external image generator like Dall-E 3. The result was that the chatbot could generate more accurate and nuanced images, including different art styles, infographics, PPTs, photorealistic images, and even edit users’ images.
Naturally, users experimented with different art styles using ChatGPT, but Ghibli-style images seem to have struck a cord among the masses.
