Probably not my best work to date, but at the time it was. The tools I used were ChatGPT and Higgsfield. I structured my prompts as if I was directing a photoshoot—starting with the realism/style foundation, then describing the subject, outfit, pose, environment, lighting, mood, camera angle, and finishing with extra detail enhancers/textures. I was basically trying to guide the AI model to define who is in the image, what they’re doing, where they are, how it’s lit, and how it should feel/look. As we know the more structured your prompt is, the more controlled and consistent your outputs become. Also, I normally upscale my images to 4K, but these weren’t.