DEV Community

Marco Rafael
Marco Rafael

Posted on

Image descriptor using gemini

Google AI Challenge Submission

This is a submission for the Google AI Studio Multimodal Challenge

What I Built

The Applet solves a chronic problem that nuts visually impaired, which is not knowing the descriptions of the images.
It uses the Gemini-2.5-flash to analyze and describe the image sent, based on defined description rules, such as not omitting opinion, for example.

Demo

https://ai-image-describer-129535786468.us-west1.run.app

How I Used Google AI Studio

I used Google Ai Studio to request the code generation of the solution.
I define description rules, requested accessibility implementation on the page and corrected bugs that emerged in development.

Multimodal Features

  • gemini-2.5-flash

Top comments (0)