Traditional Chinese Text Recognition Dataset: Synthetic Dataset and Labeled Data

Overview

Traditional Chinese Text Recognition Dataset: Synthetic Dataset and Labeled Data

Authors: Yi-Chang Chen, Yu-Chuan Chang, Yen-Cheng Chang and Yi-Ren Yeh

Paper: https://arxiv.org/abs/2111.13327

Scene text recognition (STR) has been widely studied in academia and industry. Training a text recognition model often requires a large amount of labeled data, but data labeling can be difficult, expensive, or time-consuming, especially for Traditional Chinese text recognition. To the best of our knowledge, public datasets for Traditional Chinese text recognition are lacking.

We generated over 20 million synthetic data and collected over 7,000 manually labeled data TC-STR 7k-word as the benchmark. Experimental results show that a text recognition model can achieve much better accuracy either by training from scratch with our generated synthetic data or by further fine-tuning with TC-STR 7k-word.

Synthetic Dataset: TCSynth

Inspired by MJSynth, SynthText and Belval/TextRecognitionDataGenerator, we propose a framework for generating scene text images for Traditional Chinese. To produce synthetic text images similar to real-world ones, we use different kinds of mechanisms for rendering, including word sampling, character spacing, font types/sizes, text coloring, text stroking, text skewing/distorting, background rendering, text Location and noise.

synth_text_pipeline

TCSynth dataset includes 21,535,590 synthetic text images.

TCSynth-VAL dataset includes 6,000 synthetic text images for validation.

LMDB Format

After untaring,

TCSynth/
├── data.mdb
└── lock.mdb

Our data structure of LMDB follows the repo. clovaai/deep-text-recognition-benchmark. The value queried by key 'num-samples'.encode() gets total number of text images. The indexes of text images starts from 1. Given the index, we can query binary of the image and its label by key 'image-%09d'.encode() % index and 'label-%09d'.encode() % index. The implement details are shown in the class LmdbConnector in lmdb_tools/lmdb_connector.py.

We also provide several tools to manipulate the LMDB shown in lmdb_tools. Before using those tools, we should install some dependencies. (tested with python 3.6)

pip install -r lmdb_tools/requirements.txt
  • Insert images into LMDB
python lmdb_tools/prepare_lmdb.py \
  --input_dir IMG_FOLDER \
  --gt_file GT \
  --output_dir LMDB_FOLDER
  • Insert images into LMDB (asynchronous version)
python lmdb_tools/prepare_lmdb_async.py \
  --input_dir IMG_FOLDER \
  --gt_file GT \
  --output_dir LMDB_FOLDER \
  --workers WORKERS
  • Extract images from LMDB (asynchronous version) (convert LMDB Format to Raw Format)
python lmdb_tools/extract_to_files.py \
  --input_lmdb LMDB_FOLDER \
  --output_dir IMG_FOLDER \
  --workers WORKERS

Raw Format

After untaring,

TCSynth_raw/
├── labels.txt
├── 0000/
│   ├── 00000001.jpg
│   ├── 00000002.jpg
│   ├── 00000003.jpg
│   └── ...
├── 0001/
├── 0002/
└── ...

format of labels.txt: {imagepath}\t{label}\n, for example:

0000/00000001.jpg 㒓
...

Labeled Data: TC-STR 7k-word

Our TC-STR 7k-word dataset collects about 1,554 images from Google image search to produce 7,543 cropped text images. To increase the diversity in our collected scene text images, we search for images under different scenarios and query keywords. Since the collected scene text images are to be used in evaluating text recognition performance, we manually crop text from the collected images and assign a label to each cropped text box.

TC-STR_demo

TC-STR 7k-word dataset includes a training set of 3,837 text images and a testing set of 3,706 images.

After untaring,

TC-STR/
├── train_labels.txt
├── test_labels.txt
└── images/
    ├── xxx_1.jpg
    ├── xxx_2.jpg
    ├── xxx_3.jpg
    └── ...

format of xxx_labels.txt: {imagepath}\t{label}\n, for example:

images/billboard_00000_010_雜貨鋪.jpg 雜貨鋪
images/sign_02616_999_民生路.png 民生路
...

Citation

Please consider citing this work in your publications if it helps your research.

@article{chen2021traditional,
  title={Traditional Chinese Synthetic Datasets Verified with Labeled Data for Scene Text Recognition},
  author={Yi-Chang Chen and Yu-Chuan Chang and Yen-Cheng Chang and Yi-Ren Yeh},
  journal={arXiv preprint arXiv:2111.13327},
  year={2021}
}
Owner
Yi-Chang Chen
大家好!我是YC,是一名資料科學家,熟悉機器學習和深度學習的各類技術,以及大數據分散式系統; 同時,我也是一名街頭藝人和部落客。我總是嘗試各種生命的可能性,因為我深信:人生的意義在於體驗一切身為人的經驗。
Yi-Chang Chen
This simple Python program calculates a love score based on your and your crush's full names in English

This simple Python program calculates a love score based on your and your crush's full names in English. There is no logic or reason in the calculation behind the love score. The calculation could ha

p.katekomol 1 Jan 24, 2022
Analyse japanese ebooks using MeCab to determine the difficulty level for japanese learners

japanese-ebook-analysis This aim of this project is to make analysing the contents of a japanese ebook easy and streamline the process for non-technic

Christoffer Aakre 14 Jul 23, 2022
Anuvada: Interpretable Models for NLP using PyTorch

Anuvada: Interpretable Models for NLP using PyTorch So, you want to know why your classifier arrived at a particular decision or why your flashy new d

EDGE 102 Oct 01, 2022
A repo for open resources & information for people to succeed in PhD in CS & career in AI / NLP

A repo for open resources & information for people to succeed in PhD in CS & career in AI / NLP

420 Dec 28, 2022
iBOT: Image BERT Pre-Training with Online Tokenizer

Image BERT Pre-Training with iBOT Official PyTorch implementation and pretrained models for paper iBOT: Image BERT Pre-Training with Online Tokenizer.

Bytedance Inc. 435 Jan 06, 2023
KoBERT - Korean BERT pre-trained cased (KoBERT)

KoBERT KoBERT Korean BERT pre-trained cased (KoBERT) Why'?' Training Environment Requirements How to install How to use Using with PyTorch Using with

SK T-Brain 1k Jan 02, 2023
An ActivityWatch watcher to pose questions to the user and record her answers.

aw-watcher-ask An ActivityWatch watcher to pose questions to the user and record her answers. This watcher uses Zenity to present dialog boxes to the

Bernardo Chrispim Baron 33 Dec 03, 2022
Problem: Given a nepali news find the category of the news

Classification of category of nepali news catorgory using different algorithms Problem: Multiclass Classification Approaches: TFIDF for vectorization

pudasainishushant 2 Jan 09, 2022
DziriBERT: a Pre-trained Language Model for the Algerian Dialect

DziriBERT is the first Transformer-based Language Model that has been pre-trained specifically for the Algerian Dialect.

117 Jan 07, 2023
Source code for the paper "TearingNet: Point Cloud Autoencoder to Learn Topology-Friendly Representations"

TearingNet: Point Cloud Autoencoder to Learn Topology-Friendly Representations Created by Jiahao Pang, Duanshun Li, and Dong Tian from InterDigital In

InterDigital 21 Dec 29, 2022
This code extends the neural style transfer image processing technique to video by generating smooth transitions between several reference style images

Neural Style Transfer Transition Video Processing By Brycen Westgarth and Tristan Jogminas Description This code extends the neural style transfer ima

Brycen Westgarth 110 Jan 07, 2023
RIDE automatically creates the package and boilerplate OOP Python node scripts as per your needs

RIDE: ROS IDE RIDE automatically creates the package and boilerplate OOP Python code for nodes as per your needs (RIDE is not an IDE, but even ROS isn

Jash Mota 20 Jul 14, 2022
Simple Text-To-Speech Bot For Discord

Simple Text-To-Speech Bot For Discord This is a very simple TTS bot for discord made with python. For this bot you need FFMPEG, see installation to se

1 Sep 26, 2022
The Sudachi synonym dictionary in Solar format.

solr-sudachi-synonyms The Sudachi synonym dictionary in Solar format. Summary Run a script that checks for updates to the Sudachi dictionary every hou

Karibash 3 Aug 19, 2022
State-of-the-art NLP through transformer models in a modular design and consistent APIs.

Trapper (Transformers wRAPPER) Trapper is an NLP library that aims to make it easier to train transformer based models on downstream tasks. It wraps h

Open Business Software Solutions 42 Sep 21, 2022
A Facebook Messenger Chatbot using NLP

A Facebook Messenger Chatbot using NLP This project is about creating a messenger chatbot using basic NLP techniques and models like Logistic Regressi

6 Nov 20, 2022
Conditional probing: measuring usable information beyond a baseline

Conditional probing: measuring usable information beyond a baseline

John Hewitt 20 Dec 15, 2022
Training RNNs as Fast as CNNs

News SRU++, a new SRU variant, is released. [tech report] [blog] The experimental code and SRU++ implementation are available on the dev branch which

Tao Lei 14 Dec 12, 2022
🤗Transformers: State-of-the-art Natural Language Processing for Pytorch and TensorFlow 2.0.

State-of-the-art Natural Language Processing for PyTorch and TensorFlow 2.0 🤗 Transformers provides thousands of pretrained models to perform tasks o

Hugging Face 77.3k Jan 03, 2023
Deeply Supervised, Layer-wise Prediction-aware (DSLP) Transformer for Non-autoregressive Neural Machine Translation

Non-Autoregressive Translation with Layer-Wise Prediction and Deep Supervision Training Efficiency We show the training efficiency of our DSLP model b

Chenyang Huang 37 Jan 04, 2023