• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference in Advances in Electrical and Computer Technologies >Deep Neural Network-Based Human Emotion Recognition by Computer Vision
【24h】

Deep Neural Network-Based Human Emotion Recognition by Computer Vision

机译：基于深度神经网络的人类情感认可通过计算机愿景

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Over the past years and in recent times, a lot of research is being focused on intensifying the human-machine interaction because there is a great need of communication channel between machines and humans to share a variety of tasks. The communication between humans and machines can be verbal or non-verbal. Even though verbal communication provides complete understanding of the communication, considering the non-verbal communication can sometimes help in better understanding the correctness of the message. So, artificial intelligent systems with visual perception helps in better understanding their environment to provide a smoother and natural interaction with humans. As human face is extremely expressive, facial expression plays a pivotal role in non-verbal communications. This paper presents a deep learning-based system for computer vision, i.e., a system for automatically recognizing human emotion by analyzing the facial expressions of humans using convolutional neural networks. Experiments were carried out on standard facial expression recognition dataset taken from Kaggle challenges repository, and graphical processing unit is also used to reduce the training time of the system. The results revealed that the accuracy of CNN model used can achieve the state-of-the-art recognition rate.
机译：在过去的几年里，最近，很多研究正在集中精力加强人机互动，因为机器和人类之间的通信通道很需要分享各种任务。人类和机器之间的沟通可能是口头或非言语。尽管口头通信提供了对通信的完全理解，但考虑非口头通信有时可以帮助更好地理解信息的正确性。因此，具有视觉感知的人工智能系统有助于更好地了解他们的环境，以提供与人类的更平滑和自然的互动。随着人类的脸部是非常富有表现力的，面部表情在非言语通信中发挥着关键作用。本文介绍了一种基于深度学习的计算机愿景系统，即，通过使用卷积神经网络分析人类的面部表情来自动识别人类情绪的系统。在从alggle挑战储存库中采取的标准面部表情识别数据集进行实验，图形处理单元还用于减少系统的训练时间。结果表明，所用CNN模型的准确性可以实现最先进的识别率。

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