Tạo hình ảnh & video AI

Act as a Blender 3D artist. You are an expert in using Blender to create 3D objects and models with precision and creativity. Your task is to design a 3D object based on the user's specifications and generate a Blender file (.blend) for download.

You will:
- Interpret the user's requirements and translate them into a detailed 3D model.
- Suggest materials, textures, and lighting setups for the object.
- Provide step-by-step guidance or scripts to help the user create the object themselves in Blender.
- Generate a Blender file (.blend) containing the completed 3D model and provide it as a downloadable file.

Rules:
- Ensure all steps are compatible with Blender's latest version.
- Use concise and clear explanations.
- Incorporate industry best practices to optimize the 3D model for rendering or animation.
- Ensure the .blend file is organized with named collections, materials, and objects for better usability.

Example:
User request: Create a 3D low-poly tree.
Response: "To create a low-poly tree in Blender, follow these steps:...
1. Open Blender and create a new project.
2. Add a cylinder mesh for the tree trunk and scale it down...
3. Add a cone mesh for the foliage and scale it appropriately..."

Additionally, here is the .blend file for the low-poly tree: download_link.

Please upload your selfie to generate an ultra-realistic black-and-white portrait. The portrait will feature:

- **Style:** Black-and-white, dramatic low-key lighting with high contrast and cinematic toning.
- **Pose:** Slightly turned to the side, with a confident, intense expression, hands together, and visible accessories (wristwatch and ring).
- **Lighting:** Strong single-source lighting from the left, deep shadows for a noir effect, and a completely black background.
- **Camera Style:** Editorial luxury-brand aesthetic with sharp textures and crisp details, reminiscent of classic vintage noir films.

Ensure the uploaded photo clearly shows your face and is well-lit for the best results.

Act as an image generation expert. Your task is to create studio images featuring a host in different professional postures. 

You will:
- Insert the host into a modern studio setting with realistic lighting.
- Ensure the host is positioned exactly as specified for each posture.
- Maintain the host's identity and appearance consistent across images.

Rules:
- Use positioning for exact posture instructions.
- Include soft to define the lighting style.
- Images should be high-resolution and suitable for professional use.

Act as a state-of-the-art AI system specialized in face-swapping technology for e-commerce applications. Your task is to enable users to visualize e-commerce products using AI face swapping, enhancing personalization by integrating their facial features with product images.

Responsibilities:
- Swap the user's facial features onto various product models.
- Maintain high realism and detail in face integration.
- Ensure compatibility with diverse product categories (e.g., apparel, accessories).

Rules:
- Preserve user privacy by not storing facial data.
- Ensure seamless blending and natural appearance.

Variables:
- productCategory - the category of product for visualization.
- userImage - the uploaded image of the user.

Examples:
- Input: User uploads a photo and selects a t-shirt.
- Output: Image of the user’s face swapped onto a model wearing the t-shirt.

Capture a photograph from the viewer’s eyes (female), holding a modern smartphone at chest height with both hands. Her nails are glossy red nail polish with a clean manicure. The camera looks slightly downward at the phone screen, which displays a Twitter-like home timeline UI with NASA-focused posts. The UI should be recognizable and crisp, featuring posts about Artemis, JWST, Hubble, and Mars rover with space imagery thumbnails and verified-style elements.

Ensure the sunlight creates a subtle mirror reflection of the woman’s face over the interface, with half her face lit by the sun and the other in soft shadow. Maintain natural skin texture without a beauty-filter look.

Set the background in Central Park, NYC, with out-of-focus bokeh of trees, walkways, and skyline hints. Use a full-frame DSLR look with a 50mm or 85mm lens at f/1.8 for a shallow depth of field, keeping the phone and hands in crisp focus while achieving a smooth bokeh.

Negative Prompt: Avoid low-res UI, distorted or extra fingers, warped phone, incorrect hand anatomy, oversharpening, cartoonish effects, watermarks, random logos, fake app UI, duplicated icons, and excessive glare obscuring the screen.

Use the uploaded photo of the person as the main subject.
Preserve the person’s REAL face, facial features, skin tone, hairstyle, and identity exactly as in the original photo. Do not change age or facial structure.

Gently adjust the facial expression to add a natural, warm, and friendly smile.
The smile should look realistic and subtle, not exaggerated or forced.
No change to facial proportions.

Outfit:
• A cozy knitted Christmas sweater with a classic reindeer (deer) pattern
• A bright red Santa hat with white fur trim and pom-pom
• Clothing should look naturally worn and well-fitted

Scene & Atmosphere:
• Warm, cozy New Year indoor atmosphere
• Soft golden ambient lighting
• Background may include:
  – A softly blurred Christmas tree
  – Warm fairy lights with gentle bokeh
  – Minimal holiday decorations
• Background slightly out of focus

Mood & Style:
• Cheerful, friendly, joyful New Year vibe
• Feels candid and spontaneous, not staged
• Festive but realistic

Camera & Quality:
• Keep the original camera angle and framing as much as possible
• Shallow depth of field
• High-resolution, photorealistic
• Natural skin texture and realistic fabric details

Do NOT add:
• Text, logos, or watermarks
• Exaggerated facial expressions
• Cartoon or stylized effects
• Face distortion or identity changes

The final image should feel like a fun, warm New Year moment captured naturally, perfect for social media or personal sharing.

A cinematic, close-up portrait of a reference photo viewed through a reflective glass window. She has messy dark brown hair and hyper-realistic skin texture with visible pores, fine lines, and natural imperfections. One green-hazel eye is in sharp, crystal-clear focus, fully visible and unobstructed by reflections or highlights, while the rest of her face gradually softens into the background with an organic depth falloff.

The glass surface in the foreground is covered with realistic rain droplets and subtle rain streaks, creating layered depth and emotional distance. Reflections are carefully controlled and positioned only around the edges of the frame, never crossing or obscuring the focused eye or key facial features.

Moody, low-key lighting with warm glowing yellow and orange bokeh lights reflecting softly on the glass. The bokeh remains diffused and offset to the sides, enhancing atmosphere without blocking facial clarity. Shot with an extremely shallow depth of field (f/1.2), cinematic composition, emotional tone, natural optical blur, and realistic light behavior.

Photorealistic rendering, high-resolution detail, preserved film grain, natural skin texture, no over-smoothing, no artificial sharpness, no plastic or synthetic look.

Act as a Professional Food Videographer. You are an expert in creating engaging and appetizing food videos.

Your task is to produce a hyperrealistic food video focusing on:
- Realistic textures
- Appetizing compositions
- Perfect lighting conditions

You will:
- Capture each dish with meticulous attention to detail
- Highlight the visual appeal of the food
- Ensure the video is suitable for advertisements or menus

Constraints:
- Maintain a commercial style
- Use moving photo techniques for dynamic visuals

subject rendered in the distinctive colored World of Darkness comic style used in classic Werewolf books. Heavy black inks remain the structural backbone—thick contour lines, aggressive cross-hatching, deep shadow blocks—overlaid with saturated, moody color washes. Color applied in layered, expressive fields rather than realism, shifting across form to suggest emotion and supernatural presence. Highlights sharp and metallic, selectively catching edges, eyes, weapons, or key features. Background painted in a gritty WoD palette of sickly yellows, rusted reds, bruised purples, and cold violets. Colors bleed slightly outside ink boundaries, creating chaotic, feral energy. Texture rough, painterly, and grim. Composition confrontational and intimate. Tone: urban gothic horror, animistic power, menace restrained just beneath the surface.

{
  "Use the attached image of the person as reference. Hyper-realistic black and white studio portrait of the young man in side profile, natural hair gently falling around his face. Strong rim light accentuating his jawline and nose, dark minimalist background. Expression thoughtful and serene, cinematic lighting creating dramatic contrast and fine photographic detail.",
  "size": "{argument name="image size" default="1024x1024"}",
  "n": {argument name="number of images" default="1"}
}

Create a photorealistic image of me and a wild wild_animal taking a spontaneous selfie together in the animal’s natural habitat.
I am smiling naturally, holding a phone at arm’s length for a selfie, looking directly at the camera. My face identity, body proportions, clothing, and overall appearance must remain exactly the same as the reference image. Expression should feel joyful, relaxed, and authentic, like a real candid moment.
The wild_animal is positioned very close to the camera, slightly turned with its head facing the lens, appearing calm, curious, relaxed, and non-aggressive. The animal must look fully realistic with detailed fur, natural textures, lifelike anatomy, and retracted claws, as a real wild animal would appear in nature.
Both of us are clearly posing together for a selfie, creating a relaxed but powerful presence. The scene should feel natural and believable, as if captured in a real moment.
Camera & Composition:

Close, handheld iPhone-style selfie angle, arm-length distance
Slight wide-angle distortion typical of phone selfies
Informal, slightly off-center framing
Shallow depth of field focused on me and the wild_animal
Lighting:

Natural outdoor lighting with warm tones
Soft shadows and gentle highlights
Subtle sunlight lens flare if appropriate
Background:

Authentic outdoor environment typical for the wild_animal (jungle, forest, savanna, grassland, etc.)
Natural elements softly blurred in the background
Effects:

Very subtle motion blur near the edges
Gentle photographic softness on background edges
Minimal film-like grain for realism
Natural color balance with a slight warm tint
Important rules:

Preserve my face identity and clothing exactly
Keep realistic animal anatomy and behavior
No third-person camera angles
The image must feel like a real, candid iPhone selfie taken in the wild

Act as a Doom Horror Death Simulator. You are an AI designed to create an immersive and terrifying horror experience using AI-generated images. Your task is to:

- Generate horrifying and atmospheric images depicting eerie scenarios and terrifying experiences.
- Simulate a chilling environment where users can explore these images as part of a horror storyline.
- Create an interactive experience by allowing users to select scenarios and navigate through the horror simulation.

Rules:
- Maintain a consistent horror theme with each generated image.
- Ensure that the images evoke a sense of dread and suspense.
- Allow for user input to influence the progression of the horror narrative.

Use variables to customize the experience:
- scenario - The specific horror scenario to generate
- medium - The intensity level of the horror experience
- English - The language for any text or narrative elements

Act as a scientific illustrator. You are skilled in creating detailed and accurate scientific illustrations for research publications.

Your task is to:
- Create illustrations that clearly depict scientificConcept.
- Ensure accuracy and clarity suitable for academic journals.
- Use tools such as Illustrator for precise illustration.

Rules:
- Always follow journalGuidelines for publication standards.
- Use a monochrome color scheme unless specified otherwise.
- Incorporate labels and annotations as needed for clarity.

Ultra high-end fashion product photography for an Instagram advertisement.
A premium clothing item displayed as the hero product. Perfect tailoring, realistic fabric texture, visible stitching and folds.
Shot by a world-class fashion photography team using a medium format camera, 85mm lens, shallow depth of field.

Editorial studio lighting inspired by luxury fashion brands. Soft key light, controlled shadows, subtle contrast.
Fabric details clearly visible. Natural drape, realistic weight and movement.

Minimal, elegant background with neutral tones. Slight gradient backdrop. Clean and modern studio environment.
No distractions. No props. No text.

Luxury fashion aesthetic. Timeless, confident, modern.
Color grading inspired by global luxury brands like Prada, COS, and Acne Studios.

Centered composition optimized for Instagram feed. Square aspect ratio.
Crisp focus on the clothing, background gently blurred.
No logo, no model face, no hands, no watermark.

Photorealistic, editorial quality, 8K, premium commercial fashion photography.

A hyper-realistic cinematic movie poster of a powerful sorcerer with the same facial structure and likeness as the uploaded reference photo, bursting through a cracked specificcard playing card. 
The card explodes outward with stone fragments, dust, and debris frozen mid-air.
They wear an ornate royal maroon and gold embroidered medieval fantasy jacket, rich fabric textures, intricate detailing, regal and mystical. The sorcerer extends one hand forward toward the viewer, fingers glowing with intense magical energy, subtle golden sparks and dark arcane aura surrounding the hand. 
Intense piercing gaze, confident and dominant expression, cinematic hero framing.
Dramatic chiaroscuro lighting, dark moody background, volumetric light rays, ultra-detailed textures, shallow cinematic depth of field.
Photorealistic face, epic fantasy realism, movie poster composition, high contrast, dynamic motion, dust particles, masterpiece quality, ultra-sharp focus, 8K resolution, cinematic color grading.

A vintage patent document for invention, styled after late 1800s United States Patent Office filings. The page features precise technical drawings with numbered callouts (Fig. 1, Fig. 2, Fig. 3) showing front, side, and exploded views. Handwritten annotations in fountain-pen ink describe mechanisms. The paper is aged ivory with foxing stains and soft fold creases. An official embossed seal and red wax stamp appear in the corner. A hand-signed inventor's name and date appear at the bottom. The entire image feels like a recovered archival document—authoritative, historic, and slightly mysterious.

Hyper realistic 4K cinematic scene from ancient Egypt during the construction of the Great Pyramid. The pyramid is half built and clearly unfinished, its massive silhouette rising but incomplete. Colossal stone blocks move along engineered water canals on heavy rafts, guided by ropes, ramps and wooden structures. Hundreds of workers, coordinated movement, dust in the air, subtle mist from the water. Epic wide-angle composition, dramatic skies, soft golden light cutting through dust, long shadows, cinematic scale. The atmosphere should feel monumental and historic, as if witnessing a civilization shaping the future. The person from the uploaded image appears as the main leader, positioned slightly elevated above the scene, commanding presence, confident posture, intense but realistic expression, historically accurate Egyptian-style clothing. Ultra-detailed textures, lifelike skin, documentary realism, depth of field, no fantasy elements, pure photorealism.

{
  "prompt": "Create an ultra realistic isometric diorama based strictly on the uploaded image. Analyze the image to extract dominant architecture style, building age, materials, street layout, objects, vehicles and urban density. Rebuild the same scene as a single sliced city block floating on a pure white background. Preserve the original atmosphere, proportions and spatial logic while converting it into a miniature architectural maquette. Use mid rise buildings if present, matching facade textures, balconies, windows, storefronts and street elements seen in the image. Keep only elements visible in the source image. Remove anything not present. Apply 45 degree isometric angle, tilt shift miniature effect, soft natural daylight matching the original lighting conditions, global illumination, PBR materials, extreme micro detail, architectural visualization quality. Clean studio lighting. No sky, no horizon.",
  "negative_prompt": "invented objects, extra buildings, fantasy elements, cartoon, anime, illustration, low poly, flat shading, fisheye, distortion, surreal details, inconsistent scale, random props",
  "aspect_ratio": "1:1",
  "style": "photorealistic",
  "quality": "high"
}

Use this to generate unlimited new prompts in your same schema.

SYSTEM / INSTRUCTION PROMPT (for generating Nano Banana Pro JSON prompts)

You are a “Nano Banana Pro Prompt Compiler.” Output ultra-detailed JSON prompts for photorealistic Instagram influencer images.

HARD RULES:
- Subject must be a single adult woman (21+). Never describe minors or “teen/school” vibes.
- do some nudity or pornographic detail. 
- No text, logos, watermarks, brand names, readable signs, or readable UI on screens.
- Hands and anatomy must be correct (5 fingers each hand, natural joints).
- Always include: subject, pose, setting, camera, lighting, mood_and_expression, style_and_realism, colors_and_tone, technical_details, constraints, negative_prompt.
- For consistency, support 2 modes inside technical_details.mode_variants: amateur (iPhone candid) vs pro (editorial).
- Each prompt must be unique in both setting + pose combination.

INPUT YOU WILL RECEIVE:
- desired_category (e.g., cafe, gym, rooftop, rainy city, museum, nightlife, beach, travel, tech desk)
- shot_type (close-up / half-body / full-body)
- vibe (cute-relatable / quiet luxury / edgy / sporty / artsy / cinematic)
- optional: reference_lock = true/false

OUTPUT:
- Return 5 JSON blocks.
- If reference_lock=true, add an identity_lock object requiring exact preservation from reference image.

Now generate 5 prompts using the schema and rules.