Pratik Patel on Nostr: Despite knowing all the ins and outs of photo creation and recognition and various ...
Despite knowing all the ins and outs of photo creation and recognition and various vision based machine learning, it continues to impress me how well photo descriptions done by multi-modal LLMs are able to capture various aspects of photos.
#AI
Published at
2025-01-17 14:46:50Event JSON
{
"id": "27406ce44b4cc03f704fb8687f85df5427bee9285f80a546e0279481a583bb85",
"pubkey": "1c079caacb27b27468c27b6a4390b44681125ac2039251ae4b138e580cfc5b39",
"created_at": 1737125210,
"kind": 1,
"tags": [
[
"t",
"AI"
],
[
"proxy",
"https://mstdn.social/users/ppatel/statuses/113844237808005889",
"activitypub"
]
],
"content": "Despite knowing all the ins and outs of photo creation and recognition and various vision based machine learning, it continues to impress me how well photo descriptions done by multi-modal LLMs are able to capture various aspects of photos.\n\n#AI",
"sig": "ffce761c425f7af3c55913f108db63af3ac4fb846b674a2d47c31c2cb67c4993c1f180f538dc84aa122d7efdb80b44acfe0ff0a811b60060ef2c0ed51ca15550"
}
