Text result from Training

ZJULearning / pixel_link

Implementation of our paper 'PixelLink: Detecting Scene Text via Instance Segmentation' in AAAI2018

MIT License

766 stars 254 forks source link

Open nassimaNoufail opened 5 years ago

nassimaNoufail commented 5 years ago

Does anyone knows what represents the text files resulting from the training step??