A study on object detection method from manga images using CNN
A study on object detection method from manga images using CNN
复制标题
DOI：
10.1109/iwait.2018.8369633
复制
发表时间：
2018-05
期刊：
影响因子：
--
通讯作者：
Hideaki Yanagisawa;Takuro Yamashita;Hiroshi Watanabe
中科院分区：
文献类型：
--
作者：
Hideaki Yanagisawa;Takuro Yamashita;Hiroshi Watanabe
日本漫画（manga）是世界范围内受欢迎的内容。为了从漫画图像中获取元数据，研究了漫画内容的自动识别技术。最近，卷积神经网络（CNN）已被应用于漫画图像中的对象检测。R-CNN和Fast R-CNN通过选择性搜索生成区域建议。更快的R-CNN使用称为区域提议网络（RPN）的CNN层生成它们。单次拍摄多框检测器（SSD）是最新的检测方法，可对图像中的小区域执行对象分类和框调整。这些方法对自然图像是有效的。然而，目前还不清楚这些方法是否适用于漫画图像，因为这些图像特征不同于自然图像。在本文中，我们通过比较Fast R-CNN，Faster R-CNN和SSD来检查漫画对象检测的有效性。在这里，漫画对象是面板布局、对话框、角色脸和文本。实验结果表明，Fast R-CNN对面板布局和语音气球是有效的，而Faster R-CNN对字符脸和文本是有效的。
Japanese comics (manga) are popular content worldwide. In order to acquire metadata from manga images, techniques automatic recognition of manga content have been studied. Recently, Convolutional Neural Network (CNN) has been applied to object detection in manga images. R-CNN and Fast R-CNN generate region proposals by Selective Search. Faster R-CNN generates them using CNN layers called Region Proposal Network (RPN). Single Shot MultiBox Detector (SSD), the latest detection method, performs object classification and box adjustment for small regions in an image. These methods are effective to natural images. However, it is unclear whether such methods work properly to manga images or not, since those image features are different from natural images. In this paper, we examine the effectiveness of manga object detection by comparing Fast R-CNN, Faster R-CNN, and SSD. Here, manga objects are panel layout, speech balloon, character face, and text. Experimental results show that Fast R-CNN is effective for panel layout and speech balloon, whereas Faster R-CNN is effective for character face and text.