基于深度学习的场景文本检测方法在过去几年中取得了实质性进展。然而,仍有几个问题有待解决。通常,由于CNN的有限感受野大小,长曲线文本实例往往是碎片化的。此外,使用矩形或四边形包围盒的简单表示在处理更具挑战性的任意形状的文本时显得不足。此外,文本实例的规模变化很大,这导致难以通过单个分割网络进行准确预测。为了解决这些问题,我们创新性地提出了一个两阶段分割为基础的任意文本检测器命名为\textit{NASK}(\textbf{N}需要\textbf{A} \textbf{S} econd loo \textbf{K})。具体而言,\textit{NASK}由文本实例分割网络(即\textit{TIS}(\(1^{st}\)阶段))、文本RoI池化模块和基准点表达模块(称为\textit{FOX}(\(2^{nd}\)阶段))组成。首先,\textit{TIS}进行实例分割,以获得矩形文本建议与建议的组空间和通道注意力模块(\textit{GSCA}),以增加的特征表达。然后,文本RoI池将这些矩形转换为固定大小。最后,\textit{FOX}被引入到重建文本实例与更紧密的表示使用预测的几何属性,包括文本中心线,文本行方向,字符比例尺和字符方向。在\textit{Total-Text}和\textit{SCUT-CTW1500}两个公共基准上的实验结果表明,所提出的\textit{NASK}达到了最先进的效果.
英文摘要:
Deep learning-based scene text detection methods have progressed substantially over the past years. However, there remain several problems to be solved. Generally, long curve text instances tend to be fragmented because of the limited receptive field size of CNN. Besides, simple representations using rectangle or quadrangle bounding boxes fall short when dealing with more challenging arbitrary-shaped texts. In addition, the scale of text instances varies greatly which leads to the difficulty of accurate prediction through a single segmentation network. To address these problems, we innovatively propose a two-stage segmentation based arbitrary text detector named \textit{NASK} (\textbf{N}eed \textbf{A} \textbf{S}econd loo\textbf{K}). Specifically, \textit{NASK} consists of a Text Instance Segmentation network namely \textit{TIS} (\(1^{st}\) stage), a Text RoI Pooling module and a Fiducial pOint eXpression module termed as \textit{FOX} (\(2^{nd}\) stage). Firstly, \textit{TIS} conducts instance segmentation to obtain rectangle text proposals with a proposed Group Spatial and Channel Attention module (\textit{GSCA}) to augment the feature expression. Then, Text RoI Pooling transforms these rectangles to the fixed size. Finally, \textit{FOX} is introduced to reconstruct text instances with a more tighter representation using the predicted geometrical attributes including text center line, text line orientation, character scale and character orientation. Experimental results on two public benchmarks including \textit{Total-Text} and \textit{SCUT-CTW1500} have demonstrated that the proposed \textit{NASK} achieves state-of-the-art results.
Which and Where to Focus: A Simple yet Accurate Framework for Arbitrary-Shaped Nearby Text Detection in Scene Images
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院)
Adaptive Boundary Proposal Network for Arbitrary Shape Text Detection
School of Computer and Communication Engineering, University of Science and Technology Beijing(北京科技大学计算机与通信工程学院);USTB-EEasyTech Joint Lab of Artificial Intelligence, Tencent Technology (Shenzhen) Co. Ltd(腾讯科技(深圳)有限公司北京科技大学人工智能联合实验室);Institute of Artificial Intelligence, University of Science and Technology Beijing(北京科技大学人工智能研究院)
Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network
National Key Lab for Novel Software Technology, Nanjing University(南京大学计算机软件新技术国家重点实验室);Tongji University(同济大学);Megvii (Face++) Technology Inc.(旷视科技(Face++));University of Electronic Science and Technology of China(电子科技大学);The University of Adelaide(阿德莱德大学)
Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection
School of Computer and Communication Engineering, University of Science and Technology Beijing(北京科技大学计算机与通信工程学院);USTB-EEasyTech Joint Lab of Artificial Intelligence, Tencent Technology (Shenzhen) Co. Ltd(北京科技大学-易享科技人工智能联合实验室,腾讯科技(深圳)有限公司);Institute of Artificial Intelligence, University of Science and Technology Beijing(北京科技大学人工智能研究院)
ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text (RRC-ArT)
University of Malaya(马来亚大学);South China University of Technology(华南理工大学);Baidu Inc(百度公司);Computer Vision Center, Universitat Autònoma de Barcelona(巴塞罗那自治大学计算机视觉中心)
TPSNet: Reverse Thinking of Thin Plate Splines for Arbitrary Shape Scene Text Representation
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院);Mashang Consumer Finance Co., Ltd(马上消费金融股份有限公司)
近年来,场景文本检测和识别的研究重点转向了任意形状的文本,其中文本形状表示是一个基本问题。一个理想的表示应该是紧凑的,完整的,有效的,并可重用的后续识别在我们看来。然而,以前的表述在一个或多个方面存在缺陷。薄板样条(TPS)变换在场景文本识别中取得了很大的成功。受此启发,我们重新思考了TPS的用法,并最终将TPS作为任意形状文本表示的一种精致表示。TPS表示是紧凑、完整和高效的。利用预测的TPS参数,可以直接将检测到的文本区域校正为近水平的文本区域,以辅助后续的识别。为了进一步利用TPS表示的潜力,提出了边界对齐损失。在此基础上,实现了文本检测器TPSNet,该文本检测器可以方便地扩展为文本检测器。广泛的评估和消融的几个公共基准证明了所提出的方法的有效性和优越性的文本表示和定位。特别是,TPSNet在Art数据集上实现了4.4\%的检测F-Measure改进(78.4\% vs. 74.0\%),在Total-Text上实现了5.0\%的端到端点样F-Measure改进(78.5\% vs. 73.5\%),这是大边距,没有花里胡哨的。
Learning to Predict More Accurate Text Instances for Scene Text Detection
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院)
MorphText: Deep Morphology Regularized Arbitrary-shape Scene Text Detection
Faculty of Engineering and IT, University of Technology Sydney(悉尼科技大学工程与信息技术学院);School of Computer Science and Information Security, Guilin University of Electronic Technology(桂林电子科技大学计算机科学与信息安全学院)