In this video, I show how to use Gemini pro vision on Google Cloud Vertex AI studio. I start by testing the extract text from image functionality on a picture of a book obscured by glasses. To test the Gemini pro vision functionality around understanding video I take a video of Korean pork belly on a grill and ask the model to give me some contextual information about what is happening in the video. To be honest the results blew my mind, the AI race is on between Google and Open AI.
11 сен 2024