https://github.com/FakerYFX/InceptText-Tensorflow

An Implementation of the alogrithm in paper IncepText: A New Inception-Text Module with Deformable PSROI Pooling for Multi-Oriented Scene Text Detection
https://github.com/FakerYFX/InceptText-Tensorflow

Last synced: 9 days ago
JSON representation

An Implementation of the alogrithm in paper IncepText: A New Inception-Text Module with Deformable PSROI Pooling for Multi-Oriented Scene Text Detection

• Host: GitHub
• URL: https://github.com/FakerYFX/InceptText-Tensorflow
• Owner: FakerYFX
• Created: 2018-09-06T09:41:48.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2018-11-22T14:31:42.000Z (almost 6 years ago)
• Last Synced: 2024-10-20T13:36:37.169Z (22 days ago)
• Language: Python
• Homepage:
• Size: 976 KB
• Stars: 115
• Watchers: 8
• Forks: 31
• Open Issues: 10
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# InceptText-Tensorflow

An Implementation of the alogrithm in paper IncepText: A New Inception-Text Module with Deformable PSROI Pooling for Multi-Oriented Scene Text Detection

## Introduction

### Tensorflow=1.4.0

### Preparation

1.gcc 4.9

2.cuda8.0

3.cd lib && make

- 可能遇到的错误：

1.![error1](error_pic/error1.jpg)

解决办法：把cuda路径添加到系统环境变量，然后改为#include

2.![error2](error_pic/error2.jpg)

解决办法：找到nsync_cv.h的绝对路径然后include

3.![error3](error_pic/error3.jpg)

解决办法：找到nsync_mu.h的绝对路径然后include

## Download

### 1.Models trained on ICDAR 2017

### 2.Resnet V1 50 provided by tensorflow slim[ResNet-v1](http://download.tensorflow.org/models/resnet_v1_50_2016_08_28.tar.gz)

## Train

### python train_main.py

## Test

### python test.py