Pawtrait: Pet Photo Transformation with Gemini API

[AnWo]

Heyo,

I’ve built a web app that uses Gemini to transform pet photos into artistic portraits. Looking for some feedback and tips where to take this further.

Implementation:

• Firebase - for both hosting and backend
• Frontend: React + TS
• Gemini Integration: in two places:
  1. Pet feature detection (identifying bboxes, pet type)
  2. Style transformation
• Image Processing: everything else is happening client side.

This was like a weekend project, and mostly tried to do something from A to Z, as 99% of my experiements usually never see the light of day. That meant properly setting up, domain, and adding at least some little decent UI.

Biggest challenge remains the api response (most likely hitting rate limits). Would be happy to talk to someone from Google about increasing those.

Demo:

PAWTRAIT.app

Would love to hear some feedback, especially if you have a cat or a dog. Give it a few tries, getting the gist of an animal takes a few tries.