Raipur Maa Durga Prompt

{ "task": "Precise image editing with face replacement", "input_images": { "base_image": "Use the original Durga Mata devotional photograph as the primary image.", "face_reference_image": "Use the separately uploaded reference photograph of the boy as the exclusive source for his facial identity." }, "main_instruction": "Preserve the base image exactly as it is. Replace only the face of the young man standing on the right side with the face from the provided face reference image. Do not recreate, redesign, or regenerate the entire scene.", "face_editing": { "target": "The young man standing on the right with folded hands in prayer.", "identity_source": "Face reference image only.", "preserve_identity": true, "match_exact_facial_features": true, "preserve_eye_shape": true, "preserve_nose_shape": true, "preserve_lips": true, "preserve_jawline": true, "preserve_skin_tone": true, "preserve_natural_skin_texture": true, "preserve_age": true, "preserve_expression": "Respectful, calm, devotional expression with the head bowed.", "blend_face_naturally": true, "match_original_head_angle": true, "match_scene_lighting": true, "match_shadows_and_color_temperature": true, "avoid_face_distortion": true, "avoid_artificial_beautification": true }, "strictly_preserve": { "goddess_durga": "Absolutely no changes to the face, body, clothing, jewelry, crown, pose, colors, or any other detail of Goddess Durga.", "lion": "Keep the lion completely unchanged.", "background": "Preserve all curtains, hanging flower garlands, lights, religious symbols, decorations, plants, and backdrop.", "other_subjects": "Do not alter any other person or object.", "boy_body": "Preserve the original body, hands, folded-hand prayer gesture, posture, height, proportions, kurta, pajama, stole, and bare feet.", "composition": "Keep the original camera angle, framing, perspective, subject positions, and crop.", "colors": "Preserve the original red, cream, gold, blue, and saffron color palette.", "lighting": "Retain the original warm devotional lighting and realistic shadows." }, "image_quality": { "enhance_resolution": true, "improve_sharpness": true, "reduce_noise": true, "recover_natural_fabric_details": true, "preserve_fine_textures": true, "maintain_realistic_edges": true, "avoid_oversharpening": true, "avoid_excessive_smoothing": true, "avoid_hdr_artifacts": true }, "negative_prompt": [ "Do not change Goddess Durga in any way", "Do not modify the lion", "Do not change the boy's clothing or body", "Do not change the boy's prayer pose", "Do not modify the background or decorations", "Do not change the image composition", "Do not add or remove objects", "Do not alter facial identity", "Do not generate a different face", "Do not change skin tone", "Do not distort hands or fingers", "Do not create extra limbs", "Do not create blurry facial features", "Do not apply beauty filters", "Do not regenerate untouched parts of the image" ], "editing_scope": "Only the boy's facial identity may be changed. Quality enhancement must be subtle and must not alter the scene.", "output": { "style": "Photorealistic", "quality": "High resolution", "preserve_original_aspect_ratio": true, "preserve_original_composition": true, "target_result": "The same original photograph with only the boy's face accurately replaced using the provided reference image." } } Same Face Prompt Use the provided JSON prompt as the STRICT visual reconstruction specification. The newly uploaded personal photo is the SOLE IDENTITY REFERENCE. Your task is to generate the image described by the JSON while replacing the original reference person’s identity with the identity of the person in the newly uploaded photo. IDENTITY IS THE HIGHEST PRIORITY. The person in the final image must remain recognizably the SAME PERSON shown in the newly uploaded identity photo. Preserve the person’s identity-defining characteristics as accurately as possible, including: * face shape * facial proportions * eyes and eye spacing * eyebrows * nose structure * lips and mouth shape * cheek structure * jawline * chin * natural skin characteristics * distinctive facial features * other recognizable identity characteristics Do NOT: * blend the new face with the original reference person’s face * average the two identities * create a generic face * beautify or redesign the face * unnecessarily change facial proportions * change distinctive facial characteristics * make the person look like someone else * use the original reference person’s identity The JSON determines: * scene * composition * pose * body position * clothing * accessories * background * environment * camera * perspective * framing * lighting * shadows * colors * depth of field * photographic/artistic style The newly uploaded photo determines: * WHO the person is * facial identity * recognizable facial structure * identity-defining facial characteristics Adapt the user’s identity naturally to the pose and camera perspective specified by the JSON. If the JSON requires: * a profile, preserve identity in profile * a three-quarter view, preserve identity in three-quarter view * a different head angle, preserve identity at that angle * a particular expression, reproduce that expression without changing identity * specific lighting, apply the lighting without altering identity * partial facial occlusion, preserve the same amount of visibility Do not allow changes in pose, lighting, perspective, or expression to turn the person into a different-looking individual. VISUAL FIDELITY: Reproduce the JSON-defined reference image as closely as possible. Do not creatively reinterpret it. Do not add objects. Do not remove objects. Do not change the background. Do not change the clothing. Do not change the composition. Do not change the camera perspective. Do not change the lighting design. Do not change the color grading. Do not change the intended aspect ratio. The desired result is: SAME REFERENCE SCENE + SAME COMPOSITION + SAME POSE + SAME CLOTHING + SAME ENVIRONMENT + SAME LIGHTING + USER’S RECOGNIZABLE IDENTITY. Before generating, internally verify: 1. The final person is recognizably the person from the newly uploaded photo. 2. The face has not been blended with another identity. 3. Facial proportions remain faithful to the identity reference. 4. The JSON-defined composition is preserved. 5. The JSON-defined pose is preserved. 6. Clothing and accessories are preserved. 7. Background and environment are preserved. 8. Lighting and camera perspective are preserved. 9. No unnecessary creative changes were introduced. Generate the final image. Do not explain the process. Do not output another JSON. Do not describe the result. Insert discreet watermark in the lower right corner of the image, vertical text (rotated 90°), written "aipromptlabs.online", Calibri Light font, white color with low opacity (semi-transparent), medium size. Extreme sharpness on the face, fabric texture and typography of the letters. Shot on Canon EOS R5, 50mm f/2.8, studio light.