all AI news
Image Captioners Sometimes Tell More Than Images They See. (arXiv:2305.02932v2 [cs.CV] UPDATED)
cs.CV updates on arXiv.org arxiv.org
Image captioning, a.k.a. "image-to-text," which generates descriptive text
from given images, has been rapidly developing throughout the era of deep
learning. To what extent is the information in the original image preserved in
the descriptive text generated by an image captioner? To answer that question,
we have performed experiments involving the classification of images from
descriptive text alone, without referring to the images at all, and compared
results with those from standard image-based classifiers. We have evaluate
several image captioning …
arxiv captioning deep learning generated image images image-to-text information text the information