The OCRNet model recognizes optical characters or text on images.
The training algorithm optimizes the network to minimize the connectionist temporal classification (CTC) loss between a ground truth characters sequence of a text image and a predicted characters sequence. Then characters will be decoded from the sequence output of the model through best path decoding method (greedy decoding).
The primary use case for this model is recognizing against characters or text on images.
The datasheet for the model is captured in its model card hosted at NGC.