首页> 外文期刊>Pattern recognition letters >Saliency guided deep network for weakly-supervised image segmentation
【24h】

Saliency guided deep network for weakly-supervised image segmentation

机译:显着性引导深度网络用于弱监督图像分割

获取原文
获取原文并翻译 | 示例
           

摘要

Weakly-supervised image segmentation is an important task in computer vision. A key problem is how to obtain high-quality objects location from an image-level category. Classification activation mapping is a common method which can be used to generate high-precise object location cues. However, these location cues are generally very sparse and small such that they can not provide adequate information for image segmentation. In this paper, we propose a saliency guided image segmentation network to resolve this problem. We employ a self-attention saliency method to generate subtle saliency maps and render the location cues grow as seeds by seeded region growing method to expand pixel-level labels extent. In the process of seeds growing, we use the saliency values to weight the similarity between pixels to control the growing. Therefore saliency information could help generate discriminative object regions, and the effects of wrong salient pixels can be suppressed efficiently. Experimental results on a common segmentation dataset PASCAL VOC2012 demonstrate the effectiveness of our method. (C) 2019 Elsevier B.V. All rights reserved.
机译:弱监督图像分割是计算机视觉中的重要任务。一个关键问题是如何从图像级类别获得高质量的对象位置。分类激活映射是一种常见的方法,可用于生成高精度的对象位置提示。但是,这些位置提示通常非常稀疏且很小,因此它们不能为图像分割提供足够的信息。在本文中,我们提出了一个显着性指导的图像分割网络来解决这个问题。我们采用一种自注意力显着性方法来生成微妙的显着性图,并通过种子区域生长方法来使位置提示随着种子的生长而增长,从而扩展像素级标签的范围。在种子生长的过程中,我们使用显着性值对像素之间的相似度进行加权以控制生长。因此,显着性信息可以帮助生成区分对象区域,并且可以有效地抑制错误的显着像素的影响。在通用分割数据集PASCAL VOC2012上的实验结果证明了我们方法的有效性。 (C)2019 Elsevier B.V.保留所有权利。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号