Copiedcatz

Copiedcatz is an AI-powered platform that extracts "Visual DNA" from images (lighting, composition, camera settings, etc.) and allows users to generate new variations based on that "DNA".

Comment

Copiedcatz 🐱📸

💡 Inspiration

We've all been there: you see an amazing ad, a stunning movie poster, or a perfect Instagram shot, and you think, "I wish I could make something like that, but with my product."

Current AI tools are like a slot machine. You type "cool cyberpunk city" and hope for the best. Sometimes you win, but usually, you get something random. You can't control the camera angle, the lighting, or the vibe.

We were inspired to solve this "blank canvas paralysis." We wanted to build a tool that lets you copy the recipe, not just the image. We wanted to give creators the power to look at a masterpiece and say, "I'll have what she's having," but make it their own.

🚀 What it does

Copiedcatz is a platform that lets you steal the "Visual DNA" of any image and use it to build your own.

  1. Extract: Upload any image—a photo you took, a screenshot, or a design you love. Our AI analyzes it and extracts its "DNA"—the lighting, the camera lens, the composition, and the style etc.
  2. Remix: This is where the magic happens. You don't just get random pixels; you get a "recipe" (a structured plan) you're allowed to edit without breaking the overall style. Want to change the "dog in the snow" to a "chicken in a supermarket"? Just type it. The AI keeps the professional lighting and camera angle from the original but swaps the subject.
  3. Generate: The AI cooks up your new image using the professional settings from the original.
  4. Edit & Perfect: We also added professional tools to let you tweak the focus, remove backgrounds, or upscale the quality.

⚙️ How we built it

We built Copiedcatz using a lot of "new" tech, but the real "brain" is Bria AI.

  • The Brain (Bria AI): We rely heavily on Bria's unique ability to understand images as "JSON" (a structured list of details) rather than just pixels. This allows us to separate the "Style" from the "Subject."
  • The Skeleton (Next.js): We used Next.js to build a fast, modern website that feels intuitive and easy to use.
  • The Memory (Supabase): This stores all your templates and remixes so you never lose an idea.
  • The Speed (Pusher): We use real-time technology to make the "Extraction" process feel alive, showing you progress bars as the AI decodes your image.

🚧 Challenges we ran into

  • Translating "Vibes" to Math: The hardest part was taking a user's vague idea ("Make it look cool") and translating it into the strict, precise instructions the AI needs. We had to build a special "Magic Prompt" feature that acts as a translator between human language and machine code.
  • The "Wait" Time: AI generation generally takes time (10-20 seconds). In a world of instant gratification, that feels like forever. We had to design beautiful loading screens and progress bars that explain exactly what the AI is doing (e.g., "Analyzing lighting...", "Decoding camera lens...") to keep users engaged and excited.

🏆 Accomplishments that we're proud of

  • The "Magic Prompt": We built an intelligent assistant that takes a simple sentence like "A cat in space" and automatically fills in all the complex professional photography settings (Lens type, Exposure, Lighting style) to make it look incredible.
  • True Professional Control: Unlike other tools where you just type text, we built actual sliders and buttons for things like "Depth of Field" (background blur) and "Focal Length" (zoom). It feels like holding a real camera.
  • Dynamic "No-Code" Editor: We built a smart interface that parses the raw JSON and instantly generates friendly "titled boxes" for every object. Non-tech users can simply see a box labeled "Subject 1", type "Chicken", and get a perfect result without seeing a single line of code.
  • 100% Service Coverage: We managed to integrate every single capability of the Bria AI platform—Generation, Extraction, Background Removal, and Upscaling—into one seamless app.

🧠 What we learned

We learned that structure is freedom. By forcing the AI to strictly follow a "recipe" (JSON schema), we actually give users more creative freedom. When you don't have to worry about fixing the lighting or the perspective because the template handles it for you, you can focus entirely on the fun part: the creative idea.

🔮 What's next for Copiedcatz

  • The Marketplace: We want to upgrade the app to a "Spotify for Style." Imagine browsing a library of "Visual DNAs"—one for "Moody Noir," one for "Bright Commercial," one for "Anime Action"—and applying them to your ideas with one click.
  • Mobile App: This experience would be perfect for snapping a photo of a billboard on the street and instantly remixing it on your phone.

Built With

Share this project:

Updates