Browse photo transformation examples and reusable AI image prompts for composition, lighting, color, materials, and style.
Browse a focused photo transformation category based on the PromptPlum taxonomy. Use the references as starting points, then rewrite each prompt for your own subject, format, and visual goal.
01
02
03
04Open any reference to inspect the full prompt, or take its image and aspect ratio straight into the generator.
4 public prompts01GPT Image 2 on Chatgpt Prompt: Recreate this image in a paper craft style, simplifying the details to make them suitable for paper craft artwork. Arrange the overall composition to feel visually pleasing, soft, and cute. You may add charming decorative elements such as birds, butterflies, flowers, etc., to enhance the adorable atmosphere while still matching the original image.
02A high-resolution photo of a meticulously handcrafted, 3D needle-felted wool sculpture version of the scene from the reference photo. Preserve the exact composition, poses, expressions, and number of people from the original image. Every element, including human skin, hair, and background, must be transformed into soft, dense, matte felted wool fibers with a slightly fuzzy, tangible, matte texture. Faces are simplified and stylized like plush toys with embroidered eyes and distinct blush marks on the cheeks (like in the reference photo), yet still recognizable. All clothing (including specific logos like 'VANS', specific cuts, and colors) must be rendered as knitted or felted fabric. Include realistic details typical of handmade felt toys, such as visible poke marks from the felting needle and slightly soft, rounded edges. The lighting should be soft and directional to accentuate the 3D wool textures. Shallow depth of field focusing on the felt figures against a softly blurred, felted-diorama-style background. --iw 2.0 --ar [insert aspect ratio of your photo] --stylize 150 --v 6.0
03A high-resolution photo of a meticulously handcrafted, 3D needle-felted wool sculpture version of the scene from the reference photo. Preserve the exact composition, poses, expressions, and number of people from the original image. Every element, including human skin, hair, and background, must be transformed into soft, dense, matte felted wool fibers with a slightly fuzzy, tangible, matte texture. Faces are simplified and stylized like plush toys with embroidered eyes and distinct blush marks on the cheeks (like in the reference photo), yet still recognizable. All clothing (including specific logos like 'VANS', specific cuts, and colors) must be rendered as knitted or felted fabric. Include realistic details typical of handmade felt toys, such as visible poke marks from the felting needle and slightly soft, rounded edges. The lighting should be soft and directional to accentuate the 3D wool textures. Shallow depth of field focusing on the felt figures against a softly blurred, felted-diorama-style background. --iw 2.0 --ar [insert aspect ratio of your photo] --stylize 150 --v 6.0
04Transform the uploaded photograph into a hand-painted 1990s Japanese television animation cel. Preserve the subject's identity, facial features, hairstyle, clothing, pose, composition and framing exactly. Do not redesign the subject or alter their proportions — translate the same face, the same haircut, the same outfit and the same camera angle into cel animation. The person must stay instantly recognizable. Style: hand-inked animation cel painted in gouache on acetate, photographed over a painted background plate. Bold uniform ink linework with varying weight, closed clean contours, flat cel-shaded fills with hard two-tone shadow edges — no soft gradients on skin or fabric. Lighting: single flat key with a crisp shadow terminator, small hard specular highlight blocks in the hair, warm rim light along the jaw and shoulders. Rendering: faint acetate sheen, cel dust and hairline scratches, slight registration offset between the line layer and the paint layer, 4:3-era film grain, soft analog bloom around highlights, slightly desaturated telecine color. Color palette: muted 90s broadcast tones — dusty teal, oxidized red, mustard, warm grey — skin painted in three flat steps. Background: the same setting as the photo, repainted as a soft airbrushed watercolor background plate carrying less detail than the subject, gentle depth haze. Mood: quiet, nostalgic, late-night broadcast. No 3D render, no digital vector gloss, no modern high-frame-rate shading, no HDR, no added text, no watermark, no extra characters. 👉 More looks + the app:
Choose the frame closest to your goal in composition, color, and purpose.
Change the subject, copy, style, and image ratio in the generator.
Keep the prefilled reference, generate, then refine from the result.
Start with a blank prompt and define the subject, composition, typography, and aspect ratio yourself.
Photo Transformation prompts help organize visual ideas around a specific style, subject, medium, or photography use case. Published examples may change as the collection grows.
Name the main subject first, then define composition, environment, lighting, palette, materials, camera or rendering technique, aspect ratio, and the intended output.
Remove conflicting style instructions, keep one clear visual direction, and review anatomy, text, logos, perspective, and important category-specific details before publishing.
They are reusable visual briefs grouped under the Photo Transformation category. Each brief can be adapted to a different subject, reference image, composition, or output size.
Published prompts are assigned to this collection when their subject, style, or intended use matches the Photo Transformation category.
Yes. Replace the subject and context, then adjust the lighting, palette, materials, lens or rendering method, aspect ratio, and detail level for your project.