luizgh / sigver_wiwd

Learned representation for Offline Handwritten Signature Verification. Models and code to extract features from signature images.
https://www.etsmtl.ca/Unites-de-recherche/LIVIA/Recherche-et-innovation/Projets/Signature-Verification
BSD 2-Clause "Simplified" License
138 stars 52 forks source link

signet_spp_300dpi and signet_spp_600dpi #19

Closed swagat123 closed 1 year ago

swagat123 commented 3 years ago

Hi , is there any way i can use this signet_spp_300dpi and signet_spp_600dpi using tensor flow, as i do not want to use theano. My current setup is in tensorflow.

I appreciate your suggestion. Thanks

luizgh commented 3 years ago

Hello @swagat123 . In theory you could implement this model in Tensorflow and load the weights from the Theano model. You may need to implement the SPP layer in tensorflow (although someone may have implemented it already). Anyway, you probably would still need to install Theano, to verify that you get the same results with both models

swagat123 commented 3 years ago

Thank you