fendaq Goto Github PK

text_annotation

文本标注

text_classification

all kinds of text classificaiton models and more with deep learning

text_classification-1

text_content_manipulation

Text Content Manipulation

text_deblur

Reducing the blurring noise of text images by Convolutional Neural Networks

text_detect

Text detection by pixel link and DSSD

text_detection

文本区域检测

text_detection_by_ctpn_tianchi_icpr

The Alibaba TIANCHI OCR competition

text_detection_main

text_detection_tutorial_xl

An implementation of deep learning approaches for text detection in still images and video

text_detector

Text detection model that combines Retinanet with textboxes++ for OCR

text_gcn

Graph Convolutional Networks for Text Classification. AAAI 2019

text_recognition

text_renderer

Generate text images for training deep learning ocr model

text_scalpel

Modify Chinese text, modified on LaserTagger Model. I name it "文本手术刀".目前，本项目实现了一个文本复述任务，用于NLP语料的数据增强。

text_segment_deeplearning_based

purpose This code is written by jackteng in order to understand text segment based on FCN.

text_segmentation

textinfoexp

自然语言处理相关实验（基于sougou数据集），包含文本特征提取（TF-IDF），文本分类，文本聚类，word2vec训练词向量及同义词词林中文词语相似度计算、文档自动摘要，信息抽取，情感分析与观点挖掘等。

textline-segmentation-using-fcn

textrank

基于TextRank的关键词、短语、摘要提取程序，代码使用Scala编写。

textrank4zh

:deciduous_tree:从中文文本中自动提取关键词和摘要

textrecognitiondatagenerator

A synthetic data generator for text recognition

textrotcorrect

DFT-based text image rotation correction using OpenCV

textrsn

Paper ID 336: Regression Segmentation Network for Arbitrary Shape Text Detection

textsnake.pytorch

A PyTorch implement of ECCV2018 Paper: TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes

textspotter

textsr

textsum-gan

Tensorflow re-implementation of "Generative Adversarial Network for Abstractive Text Summarization" (AAAI-18)

texttopicnet

Self-supervised learning of visual features through embedding images into text topic spaces

texture-vs-shape

Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness" (ICLR 2019 Oral)