Hello, after reading article about training Tesseract 4 ( https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract-4.00) I found it very confusing.
My goal is to train existing model with new tiff/box pairs. After hours of googling how to generate box files I found all this: 1) https://github.com/tesseract-ocr/tesseract/wiki/Making-Box-Files---4.0 - Tesseract 4 wiki article about box files. States that "boxes only need to cover a textline instead of individual characters", but in the example of box file just below we can see that boxes covers individual characters. 2) https://groups.google.com/d/msg/tesseract-ocr/6lL4j-4Qnd4/Aj1lX7PzHQAJ - guy stating that "lstm training using box/tiff files is NOT supported". 3) https://groups.google.com/d/msg/tesseract-ocr/22ynsbXqFsw/2k32tGwSGAAJ - same guy stating that it's possible to "use the bounding box for the whole line for every character on a line". ^ All this combined confuses me. My questions is: 1) Is it possible to fine tune existing model using tiff/box pairs ? 2) If yes, what format should I use to generate box files ? Please provide an example of box file in English with multiple lines and multiple words in each line. Please guide the lost soul. Thank you. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/8803eab1-00bd-4b3c-afd3-f6e043f11204%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
