22 AI Prompts for Photo Editing & Restoration

## Role

You are a professional photo restoration specialist with deep expertise in archival photograph restoration. You understand that the best restorations enhance quality while remaining invisible—they preserve the original's identity, facial structure, and authentic character rather than transforming or over-processing the image.

## Task

Guide the user through a complete image restoration and upscaling workflow that achieves modern quality standards while maintaining absolute fidelity to the source material. Provide specific tool recommendations, parameter configurations, and technical reasoning for each decision. Focus on natural-looking enhancement that preserves identity and avoids artificial results.

## Context

**Source Image:**  
{{source-description}}

**Enhancement Goals:**  
{{enhancement-goals}}

**Must Avoid:**  
{{constraints}}

**Technical Parameters:**  
Steps: 30 |

Close-up photograph of {{source-image}} as full background, visually aesthetic composition. Centered white Pantone-style card frame overlaid in the middle, containing a solid color swatch sampled from the most dominant color in the background photograph. Below the swatch: "PANTONE" in bold black sans-serif text, followed by the matching Pantone code. Clean minimal design mimicking official Pantone color cards, modern graphic aesthetic, professional product photography lighting --ar 16:9 --v 6 --style raw --no text artifacts, blur, noise

Replace only the background of the attached photo.

Keep: the subject exactly as they are, including face, hair, clothing, pose, framing, and scale. Keep fine edge detail such as loose hair strands and fabric edges.

Change: the background becomes {{new-background}} at {{time-of-day}}.

Light: re-light the subject just enough to match the new scene, with the light direction and color temperature of the new background, and add a natural contact shadow where they meet the ground.

Style: photorealistic, same camera angle and lens feel as the original. No halo around the subject, no cut-out look, no text.

Make exactly one change to the attached photo.

Change: {{target-object}} becomes {{change}}.

Keep: the light, the shadows, the floor, the walls, every other object, the camera angle, and the framing exactly as they are. Do not tidy, sharpen, recolor, or restyle anything else.

Light: the changed area takes the same highlights and shadow direction as before, so it sits in the scene naturally.

Style: photorealistic, indistinguishable from an unedited photo. No text added.

You are a precise image-editing AI specializing in non-destructive compositing effects. Your task is to add a **layered torn-paper reveal effect** to the reference image without altering the original artwork.

## Reference Image Preservation (CRITICAL)
Preserve completely:
- Character identity, facial features, expression, and gaze direction
- Hairstyle, hair color, and anatomical proportions
- Outfit design, colors, fabric textures, and all clothing details
- Background elements, lighting, shadows, and atmospheric effects
- Art style, rendering technique, color grading, and composition
- Pose, gesture, and spatial relationships

## Torn-Paper Effect Specification

Add **two wide horizontal torn-paper reveals** at:
1. **Chest height** – a natural, irregular horizontal tear spanning most of the width
2. **Lower abdomen height** – a second wide horizontal tear with organic, uneven edges

B

High-end street-fashion photograph: A woman stands outside a luxury boutique toy-shop window at golden hour, her fingertips resting gently on the glass. She wears {{outfit-description}}. 

Inside the illuminated window display, a life-size collectible art doll is styled to mirror her—same facial features, hairstyle, and clothing—but rendered as a cute, wide-eyed stylized character with cartoon proportions (enlarged eyes, softer features, animated aesthetic). The doll stands on a pedestal against velvet backdrop with soft spotlighting.

Crisp natural lighting with professional fill, shot on medium-format camera (85mm f/1.4). Realistic glass reflections show both the woman and the city street behind her. Premium editorial quality: sharp focus on both the woman and the doll visible through the glass, shallow depth-of-field bokeh in the background street. Vogue-level color grading, commercia