Open yeonsikch opened 2 years ago
this is the weight file. https://drive.google.com/file/d/16dl08NtuXX0_8N-vi6uNqQHe83eIvy3U/view?usp=sharing
Hello @yeonsikch . The ICDAR13 model is first trained on the SynthText dataset using strong supervision and later fine-tuned using weak supervision on the ICDAR 2013 dataset. As the SynthText dataset has a lot of diversity in the background and font of the text, it is possible that the model generalizes for Korean data as well.
Hi. Thank you for your successful project.
I downloaded the ICDAR2013 pre-trained file of Weakly supervised learning and made inferences.
However, it showed better performance than the MLT pre-trained data for the Korean dataset.(Even for Korean data not included in ICDAR2013) Is right the weight file learned from scratch with ICDAR 2013 data?
Thnak you.