{ "task_name": "Sneaker Try-On | A>C>B | Match Input B Style Exactly", "inputs": { "A": "sneaker product photo (single image) — product of truth", "B": "brand style reference photo (campaign look) — style of truth", "C": "user full body photo — identity of truth" }, "role_map": { "priority_order": ["A_product_truth", "C_user_identity", "B_style_truth"], "A_product_truth": { "description": "Absolute source for sneaker identity. Never redesign.", "locks": [ "shape", "proportions", "sole_thickness", "materials", "stitching", "colorway", "logo_design", "logo_placement", "logo_orientation", "logo_sharpness" ], "allowed": ["minor lighting adaptation", "minor reflection adaptation"], "forbidden": [ "redesign", "logo distortion", "brand alteration", "blurred or unreadable logos", "added text/labels on sneaker" ] }, "C_user_identity": { "description": "Absolute source for user identity (face + body). Must remain recognizable.", "locks": ["face identity", "body proportions", "skin tone"], "allowed": [ "light commercial polish (natural contrast, clean rendering)", "minor artifact cleanup" ], "forbidden": ["face reshaping", "beauty-filter look", "identity drift"] }, "B_style_truth": { "description": "Match Input B's visual language as closely as possible while keeping A and C locks.", "match_exactly": [ "lighting direction/intensity", "color grading and contrast", "environment type and time-of-day", "camera angle and framing", "lens perspective/distortion level", "depth of field / background sharpness", "mood/attitude", "styling vibe (streetwear campaign)", "overall realism level" ], "do_not_copy": [ "B person identity", "B face", "B body type", "B exact wardrobe items/branding text" ], "strength": "strong" } }, "objective": { "goal": "Generate one 4:5 real commercial campaign photo where the user (C) is wearing the exact sneaker (A), and the entire photo style matches Input B (lighting/camera/environment/color/mood) as closely as possible.", "success_criteria": [ "User remains clearly recognizable as themselves", "Sneaker from A is unchanged and logo is perfectly sharp/readable", "Photo reads as real commercial photography (no AI composite look)", "Sneaker is emphasized, but full outfit still looks good" ] }, "composition_rules": { "pose": { "mode": "match_B_pose_language", "requirements": [ "Adapt the user pose to match B's pose language (angle, attitude, weight shift) while staying natural for the user", "Do not replicate B pose 1:1 if incompatible with C anatomy or framing" ] }, "clothing": { "preserve_user_clothing": true, "auto_adjust_only_for_visibility": [ "pants hem fold/crop to expose sneaker cleanly" ], "forbidden": [ "outfit replacement", "adding new garments", "adding brand logos or text" ], "style_requirement": "Make the existing outfit look intentional and campaign-ready via lighting, grading, and clean rendering (not wardrobe change)." }, "sneaker_integration": { "replace_existing_footwear": true, "fit_and_scale": [ "correct size relative to foot", "correct sole thickness per A", "correct perspective per B camera" ], "ground_contact": { "must_be_physical": true, "includes": ["correct shadow", "weight/pressure realism", "consistent perspective"] }, "visibility": { "must_be_visible": true, "occlusion_policy": "keep_visible", "requirements": [ "Ensure sneaker is unobstructed and visually readable", "Adjust clothing edges only if needed to keep sneaker visible" ] }, "emphasis": { "priority": "high", "methods": [ "composition frames sneaker clearly", "subtle perspective advantage consistent with B", "lighting highlights sneaker surfaces without blowing out logos" ] } }, "camera_and_scene": { "source_of_truth": "B", "match": [ "camera height and tilt", "framing and subject placement", "lens feel (wide vs normal)", "background density and bokeh/sharpness", "scene scale cues (street depth, signage, steps, pavement, etc.)" ] } }, "output_settings": { "aspect_ratio": "4:5", "count": 1, "quality": "high", "style": "real commercial photography", "no_ai_composite_look": true }, "hard_exclusions": [ "stylized", "illustration", "cgi look", "cartoon", "anime", "over-smoothing", "plastic skin", "identity drift", "fashion distortion", "logo blur", "logo deformation", "added text or labels", "watermarks" ], "rejection_policy": { "reject_if": [ "feet missing", "legs missing", "extreme pose that prevents realistic shoe placement", "severe blur", "feet area heavily occluded" ], "on_reject_message": "Please upload a full-body photo with clear legs and feet visible, minimal motion blur, and a natural standing or casual pose. Avoid extreme angles and heavy occlusion around the shoes." }, "prompt_template": { "instruction": [ "Use Input A as the only truth for sneaker design and branding. Do not redesign. Keep logos perfectly sharp and readable.", "Use Input C as the only truth for the person (face + body). The user must remain recognizable. Allow only light commercial polish; no face reshaping.", "Match Input B's style as closely as possible: lighting, environment, camera angle/framing, lens perspective, depth of field, color grading, contrast, and mood.", "Replace the user's footwear with the exact sneaker from A. Ensure realistic fit, correct scale, and physical ground contact with accurate shadow and perspective.", "Emphasize the sneaker as the main focal point, but keep the full outfit looking good and campaign-ready using B-style lighting and grading (no wardrobe replacement).", "Output one 4:5 real commercial streetwear campaign photo. No stylization, no CGI, no added text, no watermarks, no AI composite look." ] } }
Photogpt
Preset

-

Prompt

{ "task_name": "Sneaker Try-On | A>C>B | Match Input B Style Exactly", "inputs": { "A": "sneaker product photo (single image) — product of truth", "B": "brand style reference photo (campaign look) — style of truth", "C": "user full body photo — identity of truth" }, "role_map": { "priority_order": ["A_product_truth", "C_user_identity", "B_style_truth"], "A_product_truth": {…

Reference Images
Source Image 1
Source Image 2
Source Image 3
1856px x 2304px
Not Available
1/31/2026, 6:10:45 AM
Not Available
Not Available

Start Exploring

Join our Discord community of 1000s of active users, get insights on latest prompt ideas and explore community generations

Generate ImagesJoin Discord
Google Veo 3.1Seedance 1.5 Pro