General Detection-based Text Line Recognition (DTLR)
Disclaimer
This repository is not affiliated with the original creators of DLTR. The model weights, code, and any other associated files are redistributed here under the terms of the Apache License 2.0 for educational and research purposes.
I have made no claim to ownership of the original model or the accompanying materials, and all credit for the development of the model and code belongs to the original authors.
For official releases and updates, please refer to the original project General Detection-based Text Line Recognition. The paper is available on arXiv and the weights can be accessed officially from Google Drive.
Checksums
These are the SHA256 checksums for the model weights
3644691dd1c37abf65b3c435897eb52aa1b426e1d7d1d033fa82c12b7974ef2f chinese/checkpoint.pth
42d610ba3ed7da720862939322c001291deb1ba9bbd76878e41ba2aa7cf7dd7c english/checkpoint.pth
ac9b4b4f1b5a54ec1d56b7845a6c1badde447dcaf1489029c171ada39b32a374 french/checkpoint.pth
36154fd678e8d5dcffa9dd69bc0a1477808a4ff4fbae7004a5da180b1ba439ac general/checkpoint.pth
840c39a672539b7a6f45b758801d668f27d95ace4f93fc334b42f74cfd6351de german/checkpoint.pth
Citation
If you find the model useful, don't forget to star the official GitHub repo :star: and cite the papers :point_down:
@article{baena2024DTLR, title={General Detection-based Text Line Recognition},
author={Raphael Baena and Syrine Kalleli and Mathieu Aubry},
booktitle={NeurIPS},
year={2024}},
url={https://arxiv.org/abs/2409.17095},