Skip to content
An Implementation of the FOTS: Fast Oriented Text Spotting with a Unified Network
C++ Python Makefile
Branch: master
Clone or download
xieyufei
xieyufei update
Latest commit 9881966 Nov 8, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
__pycache__ update Nov 8, 2018
data update Nov 8, 2018
lanms update Nov 8, 2018
models
utils update Nov 8, 2018
README.md update Nov 8, 2018
config.py update Nov 8, 2018
config.pyc update Nov 8, 2018
eval.py update Nov 8, 2018
locality_aware_nms.py update Nov 8, 2018
loss.py update Nov 8, 2018
loss.pyc update Nov 8, 2018
main_train.py update Nov 8, 2018
requrements.txt update Nov 8, 2018

README.md

FOTS: Fast Oriented Text Spotting with a Unified Network

Introduction

This is a pytorch re-implementation of FOTS: Fast Oriented Text Spotting with a Unified Network. The features are summarized blow:

  • Only detection part is implemented.

Contents

  1. Installation
  2. Download
  3. Train
  4. Test

Installation

  1. Any version of torch version >= 0.3.1 should be ok.

Download

  1. Models trained on ICDAR 2015 (training set) + ICDAR 2017 (training set)

Train

If you want to train the model, you should provide the dataset path, in the dataset path, a separate gt text file should be provided for each image and run

python main_train.py

Test

run

python eval.py

a text file will be then written to the output path.

You can’t perform that action at this time.