Closed zhang787jun closed 3 years ago
I used the code on Openvino as a reference. I haven't seen any differences.
Wow,find another version ! paper LPRNet: License Plate Recognition via Deep Neural Networks 's basic block network architectureis show as follow ,is similar to your's small_fire_block ,except padding parts. this repo show another version model trained by torch . these model use different method to cal ctc loss
your model's network architectureis are different from paper's