Segment Anything Model and the Hard Problems of Computer Vis

Listen now (80 min) | Ep. 7: Meta open sourced a model, weights, and dataset 400x larger than the previous SOTA. Joseph introduces Computer Vision for developers and what's next after OCR and Image Segmentation are solved.

Related Keywords

San Francisco , California , United States , Taiwan , Canada , Miami , Florida , Iowa , Delaware , Des Moines , China , United Kingdom , Taipei , T Ai Pei , Scotland , Iowans , Greg Brockman , Joseph Nelson , Fang Zang , Ben Silverman , Jeff Dean , Joseph Nelsonwho , Papa John , Chatgpt Plugins Hackathon , Shawn Pardo , Beyang Liu , Jacob Sorowitz , Elon Musk , George Washington , Brad Mc , Joseph Edmond , Dan Gross , Joseph Redman , Evan Spiegel , Joseph Redmond , John Papa , Google Joseph Redman , Twitter , Pinterest , Google , Cardinal Health , Youtube , Walmart , Microsoft , Computer Vision Research , University Of Washington , Us Congress , Facebook , World Knowledge Of Foundation Models , Latent Space , Computer Vision , Segment Anything Model , Zero Shot , Model Assisted , Source Graph , Magic Sudoku , Techcrunch Disrupt , Pony Express , Delaware Sea , Product Hunt Air , Hunt Air , Golden Kitty , Crunch Disrupt , Disrupt Hackathon , Board Boss , Robo Flow , Fortune One , Rob Flow , Dalle Hackathon , Glenn Yer , Chinese Google , Segment Everything , Robo Post Sam , Ask Robo , Stack Overflow , Stack Robo ,