Can I use this method to detect Document text image

ZJULearning / pixel_link

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

MIT License

766 stars 254 forks source link

Can I use this method to detect Document text image #95

Open Dgs29 opened 5 years ago

Dgs29 commented 5 years ago

Can I use implement this method for detecting text in document images, which has multiple lines of text. If yes, what changes should I make in the code.

nassimaNoufail commented 5 years ago

Same question !!

austingg commented 5 years ago

nothing need to change. just replace the dataset with document images

Dgs29 commented 5 years ago

@austingg Thanks. It worked. But I want to detect multiple lines of text in one word for a new dataset. What changes should I make to accordingly. Some like this. capture How should I train to achieve this

Dgs29 commented 5 years ago

@austingg also Is there any possible way regarding how to assign labels like Date, Title, etc to it?