Last Week in AI cover image

#156 - OpenAI's Sora, Gemini 1.5, BioMistral, V-JEPA, AI Task Force, Fun!

Last Week in AI

NOTE

Harnessing Transformer for Video Understanding

Using transformer technology traditionally used for text generation, OpenAI is innovatively employing it for video processing. By embedding images into a latent space using an encoder, they extract the meaning of each image by creating a list of numbers. This is expanded by breaking down images into patches and tracking these patches over time across video frames, enabling a deeper understanding of video content.

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner