3 ms·
That's from the "Nearest Caption in the Training Dataset". Which means it found the most similar image, and that image had that caption.
by Houshalter 11y ago
That's from the "Nearest Caption in the Training Dataset". Which means it found the most similar image, and that image had that caption.
- ilija139 11y agoNo. That is not how it works. Read the papers again.
- Houshalter 11y agoIt very clearly says "Nearest Caption in the Training Dataset". The generated labels are below it.
- mig39 11y agoWhat papers?