How good would a CBIR system trained on a dataset, for example, DELF, trained on the Google Landmarks dataset, perform when evaluated on a contextually different dataset such as the WANG or the COREL dataset without retraining?
$\begingroup$
$\endgroup$
1
-
$\begingroup$ Hi. You say "trained on a dataset, for example, DELF, trained on the Google Landmarks dataset", but a dataset cannot be trained on another dataset. Please, fix the wording of your post and add links to the relevant datasets and models. $\endgroup$– nbroSep 13, 2019 at 12:51
Add a comment
|