Back to explore9 copiesUpdated 16 days ago
Old Photo Recovery
general
advanced
Ultra-Realistic Vintage Wedding Photo Restoration
Restore an old wedding photograph while preserving the bride’s original identity, saree, jewelry, veil and pose exactly, with enhanced details, natural colors, cinematic lighting and premium DSLR quality.
0.0 (0)

Prompt content
Restore and enhance this old wedding photograph while preserving the person's original identity exactly. DO NOT change the face, eyes, nose, lips, eyebrows, hairstyle, skin tone, facial expression, age, body shape, jewelry, saree, veil, or pose. Remove blur, noise, faded colors, scratches, and low-resolution artifacts. Recover fine facial details naturally without altering identity. Enhance the silk saree texture, gold jewelry details, and bridal accessories. Apply professional DSLR photography quality with ultra-sharp focus, realistic skin texture, balanced contrast, natural colors, soft cinematic lighting, and high dynamic range. Use professional wedding portrait lighting with soft key light, gentle fill light, subtle rim light, and realistic shadows. Maintain the original outdoor background while slightly softening it with a natural DSLR bokeh effect. Improve depth, clarity, color grading, and overall image quality. Make it look as if it was captured today using a Canon EOS R5 or Sony A7R V with an 85mm f/1.4 portrait lens. Ultra realistic, photorealistic, 8K, HDR, RAW quality, professional wedding photography, extremely detailed, natural skin texture, true-to-life colors. Negative Prompt: Do not change the person's identity, face, eyes, nose, lips, hairstyle, expression, skin tone, body shape, jewelry, saree color, veil, pose, hands, fingers, or background composition. No beauty filter, no AI face replacement, no cartoon effect, no painting, no CGI, no plastic skin, no oversmoothing, no extra people, no extra limbs, no watermark, no text.
Reviews (0)
Leave a review
No reviews yet
Be the first to share your experience and tips for getting the best output with this prompt.



