【24h】

Deep Learning for People Counting Model

机译:对人们计算模型的深度学习

获取原文

摘要

Modeling of automatic people detection and counting in a real-time video is an important feature in a smart surveillance system for safety and security management, marketing research, etc. Face recognition is one of the methods which is used for people detection. In this paper, a real-time automated model is designed using deep learning algorithm such as convolutional neural network which is computationally efficient. The face detected using the proposed algorithm exploits the challenges such as variations in size and shape of the head region to achieve robust detection of a human, even under partial occlusion, dynamically changing background, and varying illumination condition. Here, we have used WIDER face dataset and FDDB dataset to show the results of the proposed method.
机译:在实时视频中自动人士检测和计数的建模是安全和安全管理,营销研究等智能监控系统中的一个重要特征。人脸识别是用于人们检测的方法之一。在本文中,使用深度学习算法设计了实时自动模型,例如卷积神经网络,该卷积神经网络是计算效率的。使用所提出的算法检测的面部利用头部区域的尺寸和形状的诸如变化的挑战,以实现人类的鲁棒检测,即使在部分闭塞,动态地改变背景和变化的照明条件。在这里,我们使用了更广泛的面部数据集和FDDB数据集来显示所提出的方法的结果。

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