That's actually I very interesting question, I am investigating. | by Ahmed Taha | Medium

Ahmed Taha

Mar 16, 2019·1 min read

That's actually I very interesting question, I am investigating.

In Defense of the Triplet Loss for Visual Recognition

We employ triplet loss as a space embedding regularizer to boost classification performance. Standard architectures…

arxiv.org

I have better results and faster implementation I am planning to submit.

Ahmed Taha

I write reviews on computer vision papers. Writing tips are welcomed.

Why they didn’t report the AUC and MAP for “triplet + softmax” in table 1?

Hosein Hasani

More from Ahmed Taha

I write reviews on computer vision papers. Writing tips are welcomed.

Get the Medium app