For the complete documentation index, see llms.txt. This page is also available as Markdown.

💡影片數據如何為多模態訓練做結構化?

影片附帶橫跨影片、音訊、字幕、場景與時序訊號的結構化對齊,而非僅僅原始檔案。

開箱即支援動作識別、內容理解、影片檢索與影片語言模型。

Last updated

Was this helpful?