how to perform four var/invar occsions with one image input ?

rpautrat / LISRD

Local Invariance Selection at Runtime for Descriptors (LISRD)

MIT License

256 stars 19 forks source link

how to perform four var/invar occsions with one image input ? #4

Closed xflv closed 4 years ago

xflv commented 4 years ago

the four variance/invariance occasions is the same in your defination, I 'm confused , how to perform four var/invar occsions with one image input ? thank you.

xflv commented 4 years ago

and where is the homography? thanks again!

rpautrat commented 4 years ago

Hi,

Yes the definition is the same for the four variances/invariances, but what differs between them is the loss function used for them: invar_desc_triplet_loss for invariant ones and var_desc_triplet_loss for the others.

The homography is in the 'inputs' dictionary: you can access it with inputs['homography'].

rpautrat commented 4 years ago

I am closing this now as the question has been answered. Feel free to reopen it if needed.