...
首页> 外文期刊>Pattern Analysis and Applications >An approach to improve SSD through mask prediction of multi-scale feature maps
【24h】

An approach to improve SSD through mask prediction of multi-scale feature maps

机译:通过多尺度特征映射的掩模预测改进SSD的方法

获取原文
获取原文并翻译 | 示例
           

摘要

We propose a novel single shot object detection network with a mask prediction branch. Our motivation is to enhance object detection features with semantic information extracted from deeper layers. The proposed mask prediction branch enriches important features in shallower layers with pixel-wise probability distribution of semantic information. Meanwhile, an improved receptive field block is adopted to increase the scale of receptive field of backbone network without too much extra computing burden. Our network improves the performance significantly over SSD and FSSD (Feature Fusion Single Shot Multi-box Detector) with just a little speed drop. In addition, we discuss the relationship between effective receptive fields and theoretical receptive fields on VGG16 backbone network. Comprehensive experimental results on PASCAL VOC 2007 demonstrate the effectiveness of the proposed method. We achieve a mAP of 79.8 with 300 x 300 input images (81.2 mAP by 512 x 512 inputs) at the speed of 58.4 FPS on a single Nvidia 1080Ti GPU. Experimental results demonstrate that the proposed network achieves a comparable performance with the state-of-the-arts.
机译:我们提出了一种具有掩模预测分支的新型单射对象检测网络。我们的动机是增强对象检测特征,具有从更深层层提取的语义信息。所提出的掩模预测分支在较浅层中丰富了具有像素的概念性分布的较浅的概率分布。同时,采用改进的接收场块来增加骨干网的接受领域的规模,而不会额外的计算负担。我们的网络通过SSD和FSSD(特征融合单次盒子多箱探测器)显着提高了性能,只需一点速度。此外,我们讨论了VGG16骨干网上有效接受领域与理论接受领域的关系。 Pascal VOC 2007上的综合实验结果证明了该方法的有效性。我们在单个NVIDIA 1080TI GPU上实现了300 x 300输入图像(81.2映射)的速度为300 x 300输入图像(81.2映射)。实验结果表明,该网络达到了与最先进的相当的性能。

著录项

  • 来源
    《Pattern Analysis and Applications》 |2021年第3期|1357-1366|共10页
  • 作者单位

    Nanjing Univ Posts & Telecommun Coll Automat Nanjing Peoples R China;

    Nanjing Forestry Univ Coll Mech & Elect Engn Nanjing Peoples R China;

    Nanjing Univ Posts & Telecommun Coll Automat Nanjing Peoples R China;

  • 收录信息 美国《科学引文索引》(SCI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    SSD; FPN; Softmax; Deep learning;

    机译:SSD;FPN;softmax;深度学习;

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