liaorongfan/EAST-1

Description

This is a PyTorch Re-Implementation of EAST: An Efficient and Accurate Scene Text Detector.

Only RBOX part is implemented.
Using dice loss instead of class-balanced cross-entropy loss. Some codes refer to argman/EAST and songdejia/EAST
The pre-trained model provided achieves 82.79 F-score on ICDAR 2015 Challenge 4 using only the 1000 images. see here for the detailed results.

Model	Loss	Recall	Precision	F-score
Original	CE	72.75	80.46	76.41
Re-Implement	Dice	81.27	84.36	82.79

Prerequisites

Only tested on

Anaconda3
Python 3.7.1
PyTorch 1.0.1
Shapely 1.6.4
opencv-python 4.0.0.21
lanms 1.0.2

When running the script, if some module is not installed you will see a notification and installation instructions. if you failed to install lanms, please update gcc and binutils. The update under conda environment is:

conda install -c omgarcia gcc-6
conda install -c conda-forge binutils

Installation

1. Clone the repo

git clone https://github.com/SakuraRiven/EAST.git
cd EAST

2. Data & Pre-Trained Model

Download Train and Test Data: ICDAR 2015 Challenge 4. Cut the data into four parts: train_img, train_gt, test_img, test_gt.
Download pre-trained VGG16 from PyTorch: VGG16 and our trained EAST model: EAST. Make a new folder pths and put the download pths into pths

mkdir pths
mv east_vgg16.pth vgg16_bn-6c64b313.pth pths/

Here is an example:

.
├── EAST
│   ├── evaluate
│   └── pths
└── ICDAR_2015
    ├── test_gt
    ├── test_img
    ├── train_gt
    └── train_img

Train

Modify the parameters in train.py and run:

CUDA_VISIBLE_DEVICES=0,1 python train.py

Detect

Modify the parameters in detect.py and run:

CUDA_VISIBLE_DEVICES=0 python detect.py

Evaluate

The evaluation scripts are from ICDAR Offline evaluation and have been modified to run successfully with Python 3.7.1.
Change the evaluate/gt.zip if you test on other datasets.
Modify the parameters in eval.py and run:

CUDA_VISIBLE_DEVICES=0 python eval.py

liaorongfan / EAST-1