How to send image to Google Gemini for processing?

[zmike]

How to convert the image into base64 and send it to Google Gemini API for processing?

 

[lokiys]

Hi, use API or any form Gemini has, I dont know about that, but you can convert image easily with C#

byte[] imageArray = System.IO.File.ReadAllBytes(@"image file path");
string base64ImageRepresentation = Convert.ToBase64String(imageArray);

 

[zmike]

I managed to send the photo to gemini via API, and get its output to get the x, y coordinates of the button.
But the button location seems to be wrong.

Does anyone have experience using AI to determine clicking coordinates? (do a screenshot of instance, then send this screenshot to AI to ask x, y coordinates to click)