Skip to content
ModelListedExternal
Download

Persian Cdistnet Ocr

The official code of CDistNet.

Sourcehuggingface.co/mohammadsaleh40/persian-cdistnet-ocr/tree/c6ad26f6331cda1029850c5a77f9faf2a1aae8b5
KindOtherVersionRevision c6ad26f

Model source

Kind
Other
File size
93.5 MB
Version
vRevision c6ad26f
Updated
Sep 26, 2026
Downloads
0
Tasks
Captioning
Source
Hugging Face

Source description

CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition

The official code of CDistNet.

Paper Link : Arxiv Link

What's News

  • [2023-08]🌟 Our paper is accepted by IJCV
  • [2022-01]🌟 Our code is released in github
  • [2021-11]🌟 The paper can be read in Arixv: http://arxiv.org/abs/2111.11011

pipline

To Do List

  • HA-IC13 & CA-IC13
  • Pre-train model
  • Cleaned Code
  • Document
  • Distributed Training

Two New Datasets

we test other sota method in HA-IC13 and CA-IC13 datasets.

HA_CA CDistNet has a performance advantage over other SOTA methods as the character distance increases (1-6)

HA-IC13

Method123456Code & Pretrain model
VisionLAN (ICCV 2021)93.5892.8889.9782.2672.2361.03Offical Code
ABINet (CVPR 2021 )95.9295.2291.9585.7673.7564.99Offical Code
RobustScanner* (ECCV 2020)96.1595.3393.2388.9181.1071.53--
Transformer-baseline*96.2795.4592.4286.4679.3572.46--
CDistNet96.6296.1594.2889.9683.4377.71--

CA-IC13

Method123456Code & Pretrain model
VisionLAN (ICCV 2021)94.8792.7784.0175.0364.2952.74Offical Code
ABINet (CVPR 2021 )96.6295.9287.8676.3165.4654.49Offical Code
RobustScanner* (ECCV 2020)95.2294.8785.3076.5568.3860.79--
Transformer-baseline*95.6894.4085.8875.8565.9358.58--
CDistNet96.2795.5788.4579.5870.3663.13--

Datasets

The datasets are same as ABINet

Environment

package you can find in env_cdistnet.yaml.

#Installed
conda create -n CDistNet python=3.7
conda install pytorch==1.5.1 torchvision==0.6.1 cudatoolkit=9.2 -c pytorch
pip install opencv-python mmcv notebook numpy einops tensorboardX Pillow thop timm tornado tqdm matplotlib lmdb

Pretrained Models

Get the pretrained models from BaiduNetdisk(passwd:d6jd), GoogleDrive. (We both offer training log and result.csv in same file.) The pretrained model should set in models/reconstruct_CDistNetv3_3_10

Performances of the pretrained models are summaried as follows:

comment: <> (|CDistNet(paper)|6|97.67|93.82|96.57|86.25|89.77|89.58|92.28|)

comment: <> (|CDistNet(rebuild)|4|97.43|93.51|96.37|86.03|88.68|93.4|92.57|)

Train

CUDA_VISIBLE_DEVICES=0,1,2,3 python train.py --config=configs/CDistNet_config.py

Eval

CUDA_VISIBLE_DEVICES=0 python eval.py --config=configs/CDistNet_config.py

Citation

@article{Zheng2021CDistNetPM,
  title={CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition},
  author={Tianlun Zheng and Zhineng Chen and Shancheng Fang and Hongtao Xie and Yu-Gang Jiang},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.11011}
}

Sources

1 source
Hugging FacePrimaryVersion c6ad26f6331cda1029850c5a77f9faf2a1aae8b5
huggingface.co/mohammadsaleh40/persian-cdistnet-ocr/tree/c6ad26f6331cda1029850c5a77f9faf2a1aae8b5

Verified Sep 10

Model artifacts

1 artifact

models/all_tiny_13font_persian_cdistnet_128_36/epoch7_best_acc.pth

pth · 93.5 MB · SHA-256 9b3b4e0857f0…bf19 · Hugging Face

Download

Source excerpts

2 excerpts
ReadmeSep 10huggingface.co/mohammadsaleh40/persian-cdistnet-ocr

--- datasets: - mohammadsaleh40/persian-word-ocr language: - fa pipeline_tag: image-to-text --- # CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition The official code of CDistNet. Paper Link : [Arxiv Link](http://arxiv.org/abs/2111.11011) ## What's News - [2023-08]🌟 Our paper is accepted by IJCV - [2022-01]🌟 Our code is released in github - [2021-11]🌟 The paper can be read in Arixv: http://arxiv.org/abs/2111.11011 ![pipline](./utils/fig2_00.png) ## To Do List * [x] HA-IC13 & CA-IC13 * [x] Pre-train model * [x] Cleaned Code * [ ] Document * [ ] Distributed Training ## Two New Datasets we test other sota method in HA-IC13 and CA-IC13 datasets. ![HA_CA](./utils/fig5_00.png) **CDistNet has a performance advantage over other SOTA methods as the character distance increases (1-6)** ### HA-IC13 |Method |1 | 2 | 3 | 4 | 5 | 6 | Code & Pretrain model| |- | - | - | - | - | - | - | - | |VisionLAN (ICCV 2021) | 93.58 | 92.88 | 89.97 | 82.26 | 72.23 | 61.03 | [Offical Code](https://github.com/wangyuxin87/VisionLAN)| |ABINet (CVPR 2021 ) | 95.92 |95.22 | 91.95 | 85.76 | 73.75 | 64.99 | [Offical Code](https://github.com/FangShancheng/ABINet)| |RobustScanner* (ECCV 2020) | 96.15 | 95.33 | 93.23 | 88.91 | 81.10 |71.53 | -- | | Transformer-baseline* | 96.27 | 95.45 | 92.42 | 86.46 | 79.35 | 72.46 | -- | |CDistNet |**96.62**| **96.15** | **94.28** | **89.96** | **83.43** | **77.71** | -- | ### CA-IC13 |Method |1 | 2 | 3 | 4 | 5 | 6 | Code & Pretrain model| |- | - | - | - | - | - | - | - | |VisionLAN (ICCV 2021) | 94.87 | 92.77 | 84.01 | 75.03 | 64.29 | 52.74 | [Offical Code](https://github.com/wangyuxin87/VisionLAN)| |ABINet (CVPR 2021 ) | **96.62** | **95.92** | 87.86 |76.31 | 65.46 | 54.49 | [Offical Code](https://github.com/FangShancheng/ABINet)| |RobustScanner* (ECCV 2020) | 95.22 | 94.87 | 85.30 | 76.55 | 68.38 |60.79 | -- | | Transformer-baseline* | 95.68 | 94.40 | 85.88 | 75.85 | 65.93 | 58.58 | -- | |CDistNet | 96.27 | 95.57 | **88.45** | **79.58** | **70.36** | **63.13** | -- | ## Datasets **The datasets are same as ABINet** - Training datasets 1. [MJSynth](http://www.robots.ox.ac.uk/~vgg/data/text/) (MJ): - [LMDB dataset BaiduNetdisk(passwd:n23k)](https://pan.baidu.com/s/1mgnTiyoR8f6Cm655rFI4HQ) 2. [SynthText](http://www.robots.ox.ac.uk/~vgg/data/scenetext/) (ST): - [LMDB dataset BaiduNetdisk(passwd:n23k)](https://pa...

NoteSep 10huggingface.co/mohammadsaleh40/persian-cdistnet-ocr

Source context: 0 downloads · 0 likes · Pipeline image-to-text · Repo mohammadsaleh40/persian-cdistnet-ocr