Implémentation officielle de Pytorch du détecteur de texte artisanal | Papier | Modèle pré-entraîné | Supplémentaire
Youngmin Baek, Bado Lee, Dongyoon Han, Sangdoo Yun, Hwalsuk Lee.
CLOVA AI Research, Naver Corp.
Implémentation de Pytorch pour le détecteur de texte artisanal qui détecte efficacement la zone de texte en explorant chaque région de caractère et affinité entre les caractères. La boîte de délimitation des textes est obtenue en trouvant simplement des rectangles de délimitation minimaux sur la carte binaire après la région de caractères de seuil et les scores d'affinité.

13 juin 2019 : Mise à jour initiale 20 juillet 2019 : Ajout de post-traitement pour le résultat du polygone 28 sept
pip install -r requirements.txt
Le code de formation n'est pas inclus dans ce référentiel, et nous ne pouvons pas publier le code de formation complet pour la raison IP.
| Nom du modèle | Ensembles de données utilisés | Langues | But | Lien de modèle |
|---|---|---|---|---|
| Général | SynthText, IC13, IC17 | ENG + MLT | À des fins générales | Faire un clic |
| IC15 | SynthText, IC15 | Engin | Pour IC15 seulement | Faire un clic |
| Linkrefiner | CTW1500 | - | Utilisé avec le modèle général | Faire un clic |
python test.py --trained_model=[weightfile] --test_folder=[folder path to test images]
L'image du résultat et les cartes SOCRE seront enregistrées sur ./result par défaut.
--trained_model : modèle pré-entraîné--text_threshold : seuil de confiance du texte--low_text : Score du texte bas--link_threshold : Seuil de confiance en lien--cuda : utilisez Cuda pour l'inférence (par défaut: true)--canvas_size : Taille d'image maximale pour l'inférence--mag_ratio : rapport d'agrandissement de l'image--poly : activer le résultat de type polygone--show_time : afficher le temps de traitement--test_folder : Chemin de dossier pour saisir les images--refine : Utilisez un raffineur de lien pour l'ensemble de données au niveau de la sensibilité--refiner_model : modèle de raffineur pré-entraîné @inproceedings{baek2019character,
title={Character Region Awareness for Text Detection},
author={Baek, Youngmin and Lee, Bado and Han, Dongyoon and Yun, Sangdoo and Lee, Hwalsuk},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
pages={9365--9374},
year={2019}
}
Copyright (c) 2019-present NAVER Corp.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.