Cascaded Segmentation-Detection Networks for Word-Level Text Spotting.

Proceedings of the ... International Conference on Document Analysis and Recognition. International Conference on Document Analysis and Recognition Pub Date : 2017-11-01 Epub Date: 2018-01-29 DOI:10.1109/ICDAR.2017.210

Siyang Qin, Roberto Manduchi

{"title":"Cascaded Segmentation-Detection Networks for Word-Level Text Spotting.","authors":"Siyang Qin, Roberto Manduchi","doi":"10.1109/ICDAR.2017.210","DOIUrl":null,"url":null,"abstract":"<p><p>We introduce an algorithm for word-level text spotting that is able to accurately and reliably determine the bounding regions of individual words of text \"in the wild\". Our system is formed by the cascade of two convolutional neural networks. The first network is fully convolutional and is in charge of detecting areas containing text. This results in a very reliable but possibly inaccurate segmentation of the input image. The second network (inspired by the popular YOLO architecture) analyzes each segment produced in the first stage, and predicts oriented rectangular regions containing individual words. No post-processing (e.g. text line grouping) is necessary. With execution time of 450 ms for a 1000 × 560 image on a Titan X GPU, our system achieves good performance on the ICDAR 2013, 2015 benchmarks [2], [1].</p>","PeriodicalId":90689,"journal":{"name":"Proceedings of the ... International Conference on Document Analysis and Recognition. International Conference on Document Analysis and Recognition","volume":"2017 ","pages":"1275-1282"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5858575/pdf/nihms904003.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the ... International Conference on Document Analysis and Recognition. International Conference on Document Analysis and Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDAR.2017.210","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2018/1/29 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

We introduce an algorithm for word-level text spotting that is able to accurately and reliably determine the bounding regions of individual words of text "in the wild". Our system is formed by the cascade of two convolutional neural networks. The first network is fully convolutional and is in charge of detecting areas containing text. This results in a very reliable but possibly inaccurate segmentation of the input image. The second network (inspired by the popular YOLO architecture) analyzes each segment produced in the first stage, and predicts oriented rectangular regions containing individual words. No post-processing (e.g. text line grouping) is necessary. With execution time of 450 ms for a 1000 × 560 image on a Titan X GPU, our system achieves good performance on the ICDAR 2013, 2015 benchmarks [2], [1].

Abstract Image

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

用于单词级文本定位的级联分割检测网络

我们介绍了一种用于单词级文本识别的算法，它能够准确可靠地确定 "野生 "文本中单个单词的边界区域。我们的系统由两个卷积神经网络级联而成。第一个网络是完全卷积网络，负责检测包含文本的区域。这样就能对输入图像进行非常可靠但可能不准确的分割。第二个网络（受流行的 YOLO 架构启发）分析第一阶段生成的每个片段，并预测包含单个单词的定向矩形区域。无需进行后处理（如文本行分组）。我们的系统在 Titan X GPU 上处理 1000 × 560 图像的执行时间为 450 毫秒，在 ICDAR 2013 和 2015 基准测试中取得了良好的性能[2], [1]。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Proceedings of the ... International Conference on Document Analysis and Recognition. International Conference on Document Analysis and Recognition

自引率

0.00%

发文量

期刊最新文献

Cascaded Segmentation-Detection Networks for Word-Level Text Spotting. Text Detection in Natural Scene Images by Stroke Gabor Words. A survey of table recognition Analysis and interpretation of visual saliency for document functional labeling Independent component analysis for document restoration