Data

Tools

Indexes

Applications

Experiments

Open Source Science

multi-domain-ocr

https://github.com/jiayou-chao/multi-domain-ocr

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (9.9%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

Basic Info
  • Host: GitHub
  • Owner: Jiayou-Chao
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Size: 516 KB
Statistics
  • Stars: 4
  • Watchers: 1
  • Forks: 0
  • Open Issues: 1
  • Releases: 0
Created about 2 years ago · Last pushed about 2 years ago
Metadata Files
Readme License Citation

README.md

Multi-domain-OCR

The project is the codebase for the paper "Multi-domain OCR with Meta Self-Learning" (https://arxiv.org/abs/2401.00971). The code is based on MMOCR.

Installation

The environment setup is the same as MMOCR. Alternatively you can use the setup.sh script to install the environment:

```bash

(Optional) Create a conda environment

conda create -n multi-domain-ocr python=3.10 -y conda activate multi-domain-ocr

Set up the environment

bash setup.sh ```

Prepare Dataset

The dataset used for evaluation is the open-sourced dataset MSDA (Multi-source domain adaptation dataset for text recognition). Please refer to the homepage for the download link. The dataset is in the format of tar file. Please extract the file and then use tools/dataset_converters/textrecog/lmdb_converter.py to convert the dataset to lmdb format. Assume the dataset is stored in data/Meta-SelfLearning and to be extracted to data/cache/, the following is an example of converting the syn dataset to lmdb format:

```bash

Extract the dataset

mkdir -p data/cache/ tar -xvf data/Meta-SelfLearning/syn/test_imgs.tar -C data/cache/

Convert the dataset to lmdb format

python tools/datasetconverters/textrecog/lmdbconverter.py data/Meta-SelfLearning/syn/testlabel.txt data/Meta-SelfLearning/LMDB/syn/testimgs.lmdb -i data/cache/Meta-SelfLearning/root/data/TextRecognitionDatasets/IMG/syn/test_imgs/ --label-format txt ```

Training

The training script is in tools/train.py. The following is an example of training the model on the syn dataset:

bash python tools/train.py configs/path/to/config.py

If you want to use multiple GPUs for training, use tools/dist_train.sh:

bash tools/dist_train.sh configs/configs/path/to/config.py 8 --auto-scale-lr --amp

The config files to reproduce the results in the paper are in configs/. The following is an example of training the backbone on the docu dataset:

bash tools/dist_train.sh configs/textrecog/adapter/backbone_docu.py 8 --auto-scale-lr --amp

The following is an example of training the adapter on the syn dataset:

bash tools/dist_train.sh configs/textrecog/adapter/adapter_docu_adapter_syn.py 8 --auto-scale-lr --amp

Owner

  • Login: Jiayou-Chao
  • Kind: user

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
title: "OpenMMLab Text Detection, Recognition and Understanding Toolbox"
authors:
  - name: "MMOCR Contributors"
version: 0.3.0
date-released: 2020-08-15
repository-code: "https://github.com/open-mmlab/mmocr"
license: Apache-2.0

GitHub Events

Total
Last Year

Dependencies

.circleci/docker/Dockerfile docker
  • pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
docker/Dockerfile docker
  • pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
docker/serve/Dockerfile docker
  • pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
requirements/albu.txt pypi
  • albumentations >=1.1.0
requirements/build.txt pypi
  • numpy *
  • pyclipper *
  • torch >=1.1
requirements/docs.txt pypi
  • docutils ==0.16.0
  • markdown >=3.4.0
  • myst-parser *
  • sphinx ==4.0.2
  • sphinx-tabs *
  • sphinx_copybutton *
  • sphinx_markdown_tables >=0.0.16
  • tabulate *
requirements/mminstall.txt pypi
  • mmcv >=2.0.0rc4,<2.1.0
  • mmdet >=3.0.0rc5,<3.1.0
  • mmengine >=0.7.0,<1.0.0
requirements/optional.txt pypi
requirements/readthedocs.txt pypi
  • imgaug *
  • kwarray *
  • lmdb *
  • matplotlib *
  • mmcv >=2.0.0rc1
  • mmdet >=3.0.0rc0
  • mmengine >=0.1.0
  • pyclipper *
  • rapidfuzz >=2.0.0
  • regex *
  • scikit-image *
  • scipy *
  • shapely *
  • titlecase *
  • torch *
  • torchvision *
requirements/runtime.txt pypi
  • imgaug *
  • lmdb *
  • matplotlib *
  • numpy *
  • opencv-python >=4.2.0.32,
  • pyclipper *
  • pycocotools *
  • rapidfuzz >=2.0.0
  • scikit-image *
requirements/tests.txt pypi
  • asynctest * test
  • codecov * test
  • flake8 * test
  • interrogate * test
  • isort * test
  • kwarray * test
  • lanms-neo ==1.0.2 test
  • parameterized * test
  • pytest * test
  • pytest-cov * test
  • pytest-runner * test
  • ubelt * test
  • xdoctest >=0.10.0 test
  • yapf * test
requirements.txt pypi
setup.py pypi