ViTAE-Transformer / DeepSolo

The official repo for [CVPR'23] "DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting" & [ArXiv'23] "DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting"
Other
242 stars 34 forks source link

What is the specification of annotation and how to create the dataset? #62

Open shizidushu opened 6 months ago

shizidushu commented 6 months ago

I want to create my custom dataset for training with DeepSolo. But I don't know how to do it.

I know there is a tool called PPOCRLabel provided by PaddleOCR to create label for OCR task. But I don't know how to convert the labels to the format used by DeepSolo. Is there already someone has script to do the converting?

Can you provide some information about the label format used? I want to know the specification so I may write converter by myself.

Is there any other data labelling tools (not PPOCRLabel) to be suggested?