Deep Matching Prior Network: Toward Tighter Multi-oriented Text Detection

最新推荐文章于 2024-03-23 09:59:06 发布

原创最新推荐文章于 2024-03-23 09:59:06 发布 · 553 阅读

0 ·

CC 4.0 BY-SA版权

针对传统矩形框文本检测存在的背景噪声、重叠及信息丢失等问题，本文提出一种新的卷积神经网络方法——深度匹配先验网络（DMPNet），通过四边形滑动窗口粗定位文本区域，结合蒙特卡洛方法快速准确地计算多边形面积，并设计SmoothLn损失函数适度调整预测边界框，最终实现更精确的文本定位。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

Retrospective research has only focused on using rectangular bounding box or horizontal sliding window to localize text, which may result in redundant background noise, unnecessary overlap or even information loss. To address these issues, we propose a new Convolutional Neural Networks (CNNs) based method, named Deep Matching Prior Network (DMPNet), to detect text with tighter quadrangle.
这里写图片描述

quadrangle 四边形

firstly, roughly recalling text with quadrilateral sliding window
then, using a shared Monte-Carlo method for fast and accurate computing of polygonal areas;
finely localizing text with quadrangle and design a Smooth Ln loss for
moderately adjusting the predicted bounding box.

设计多几个sliding window

这里写图片描述