Visual captions: Using large language models to augment video conferences with dynamic visuals
Google Research AI blog
JUNE 6, 2023
Or when talking about your recent family trip to San Francisco, you may want to show a photo from your personal album. corresponds to visual content of “a photo from the trip to Mexico'', a visual type of “photo”, and visual source of “personal album”. Acknowledgements This work is a collaboration across multiple teams at Google.
Let's personalize your content