This repository contains the code for the ICDAR 2021 paper, "Handwriting Recognition with Novelty" by Derek S. Prijatelj, Samuel Grieggs, Futoshi Yumoto, Eric Robertson, and Walter J. Scheirer.
The training and testing data augmentations used in the paper are provided at darpa-sail-on/writer-recognition. That additional repository also contains the baseline agent and other code used in the supplementary material.
When installing out code, we recommend using a virtual environment, such as venv or conda.
The models used in experimentation are contained within hwr_novelty
hwr_novelty
├── generate
└── models
└── losses
Install using python setup.py install
The code and scripts for the experiments are contained within experiments
experiments
├── configs
│ ├── m18_par
│ ├── paper_1
│ └── par_iam_round1
│ ├── labels
│ └── v2
│ ├── crnn
│ │ └── continue
│ └── mevm
├── crc_scripts
│ ├── mevm
│ └── paper_1
│ ├── repr
│ └── writer_id
└── research
└── par_v1
└── grieggs
configs: Configuration files for running the experimentscrc_scripts: Scripts used to run our code on our machines.research: The research experiment specific code
Install using python setup_exp.py install
Our code contributions within this repository are released under the MIT License located in LICENSE.txt
If you use our work, please use the following Bibtex to cite our paper:
@inproceedings{prijatelj_handwriting_2021,
title = {Handwriting {Recognition} with {Novelty}},
author = {Prijatelj, Derek S. and Grieggs, Samuel and Yumoto, Futoshi and Robertson, Eric and Scheirer, Walter J.},
year = {2021},
editor = {Lladós, Josep and Lopresti, Daniel and Uchida, Seiichi},
isbn = {978-3-030-86337-1},
doi = {10.1007/978-3-030-86337-1_33},
booktitle = {Document {Analysis} and {Recognition} – {ICDAR} 2021},
series = {Lecture {Notes} in {Computer} {Science}},
publisher = {Springer International Publishing},
pages = {494--509},
}