Picture for Christoph Wick

Christoph Wick

Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes

Add code
Oct 13, 2021
Figure 1 for Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes
Figure 2 for Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes
Figure 3 for Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes
Figure 4 for Rescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-Prefixes
Viaarxiv icon

Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning

Add code
Jun 15, 2021
Figure 1 for Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning
Figure 2 for Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning
Figure 3 for Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning
Figure 4 for Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning
Viaarxiv icon

Optimizing small BERTs trained for German NER

Add code
Apr 23, 2021
Figure 1 for Optimizing small BERTs trained for German NER
Figure 2 for Optimizing small BERTs trained for German NER
Figure 3 for Optimizing small BERTs trained for German NER
Figure 4 for Optimizing small BERTs trained for German NER
Viaarxiv icon

OCR4all -- An Open-Source Tool Providing a Automatic OCR Workflow for Historical Printings

Add code
Sep 09, 2019
Figure 1 for OCR4all -- An Open-Source Tool Providing a Automatic OCR Workflow for Historical Printings
Figure 2 for OCR4all -- An Open-Source Tool Providing a Automatic OCR Workflow for Historical Printings
Figure 3 for OCR4all -- An Open-Source Tool Providing a Automatic OCR Workflow for Historical Printings
Figure 4 for OCR4all -- An Open-Source Tool Providing a Automatic OCR Workflow for Historical Printings
Viaarxiv icon

State of the Art Optical Character Recognition of 19th Century Fraktur Scripts using Open Source Engines

Add code
Oct 08, 2018
Viaarxiv icon

Calamari - A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognition

Add code
Aug 06, 2018
Figure 1 for Calamari - A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognition
Figure 2 for Calamari - A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognition
Figure 3 for Calamari - A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognition
Figure 4 for Calamari - A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognition
Viaarxiv icon

Improving OCR Accuracy on Early Printed Books by combining Pretraining, Voting, and Active Learning

Add code
Feb 28, 2018
Figure 1 for Improving OCR Accuracy on Early Printed Books by combining Pretraining, Voting, and Active Learning
Figure 2 for Improving OCR Accuracy on Early Printed Books by combining Pretraining, Voting, and Active Learning
Figure 3 for Improving OCR Accuracy on Early Printed Books by combining Pretraining, Voting, and Active Learning
Figure 4 for Improving OCR Accuracy on Early Printed Books by combining Pretraining, Voting, and Active Learning
Viaarxiv icon

Improving OCR Accuracy on Early Printed Books using Deep Convolutional Networks

Add code
Feb 27, 2018
Figure 1 for Improving OCR Accuracy on Early Printed Books using Deep Convolutional Networks
Figure 2 for Improving OCR Accuracy on Early Printed Books using Deep Convolutional Networks
Figure 3 for Improving OCR Accuracy on Early Printed Books using Deep Convolutional Networks
Figure 4 for Improving OCR Accuracy on Early Printed Books using Deep Convolutional Networks
Viaarxiv icon

Fully Convolutional Neural Networks for Page Segmentation of Historical Document Images

Add code
Feb 15, 2018
Figure 1 for Fully Convolutional Neural Networks for Page Segmentation of Historical Document Images
Figure 2 for Fully Convolutional Neural Networks for Page Segmentation of Historical Document Images
Figure 3 for Fully Convolutional Neural Networks for Page Segmentation of Historical Document Images
Figure 4 for Fully Convolutional Neural Networks for Page Segmentation of Historical Document Images
Viaarxiv icon

Transfer Learning for OCRopus Model Training on Early Printed Books

Add code
Dec 21, 2017
Figure 1 for Transfer Learning for OCRopus Model Training on Early Printed Books
Figure 2 for Transfer Learning for OCRopus Model Training on Early Printed Books
Figure 3 for Transfer Learning for OCRopus Model Training on Early Printed Books
Figure 4 for Transfer Learning for OCRopus Model Training on Early Printed Books
Viaarxiv icon