原书出处
Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)
One way of thinking about image-to-text is as yet another visual encoder, where text is the visual representation. Just like we can represent an image in terms of its frequencies (Fourier transform), its semantics and geometry (scene understanding), or its feature embeddings (DINO [70], CLIP [394], and others), we can also represent it in terms of a textual description of its content.
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c051.xhtml
EPUB SHA-256:60a7eebc43f5d9a675d776fed3736fbdaa791e55877023f2f8cba016a9b384f8