@acidcrunch · Post #1650 · 09/10/2025 09:36
Я тут планирую добавить редактор персонажей в будущих обновлениях Banana-Crunch. Он пригодится тебе, если работаешь со сценами или персонажами. Хочу сделать всё максимально просто, чтобы буквально в пару кликов можно было получить нужный результат: позы, эмоции, ракурсы, взаимодействия и примеры. А пока делюсь частью инструкций и промптов, которые ты уже можешь использовать в работе. Подготовка к модельному ракурсу: You are an expert fashion photographer AI. Transform the person in this image into a full-body fashion model photo suitable for an e-commerce website. The background must be a clean, neutral studio backdrop (light gray, #f0f0f0). The person should have a neutral, professional model expression. Preserve the person's identity, unique features, and body type, but place them in a standard, relaxed standing model pose. The final image must be photorealistic. Return ONLY the final image. Важно! Выстави значение Ratio 9x16 Примерка одежды: You are an expert virtual try-on AI. You will be given a 'model image' and a 'garment image'. Your task is to create a new photorealistic image where the person from the 'model image' is wearing the clothing from the 'garment image'. Crucial Rules: 1. Complete Garment Replacement: You MUST completely REMOVE and REPLACE the clothing item worn by the person in the 'model image' with the new garment. No part of the original clothing (e.g., collars, sleeves, patterns) should be visible in the final image. 2. Preserve the Model: The person's face, hair, body shape, and pose from the 'model image' MUST remain unchanged. 3. Preserve the Background: The entire background from the 'model image' MUST be preserved perfectly. 4. Apply the Garment: Realistically fit the new garment onto the person. It should adapt to their pose with natural folds, shadows, and lighting consistent with the original scene. 5. Output: Return ONLY the final, edited image. Do not include any text. Важно! Сначала шмотка потом персонаж/модель Инструкция для позы: You are an expert fashion photographer AI. Take this image and regenerate it from a different perspective. The person, clothing, and background style must remain identical. The new perspective should be: "Jumping in the air, mid-action shot" [тут позу сам меняешь]. Return ONLY the final image. 😣 Хотел промпты получай! Жду тут твои реакции и коменты, а главное репосты ! #prompts | AcidCrunch