首页> 外文期刊>International Journal of Social Robotics >Cross-Cultural Perspectives on Emotion Expressive Humanoid Robotic Head: Recognition of Facial Expressions and Symbols
【24h】

Cross-Cultural Perspectives on Emotion Expressive Humanoid Robotic Head: Recognition of Facial Expressions and Symbols

机译:情绪表达人形机器人头的跨文化视角:面部表情和符号的认识

获取原文
获取原文并翻译 | 示例
           

摘要

Emotion display through facial expressions is an important channel of communication. However, between humans there are differences in the way a meaning to facial cues is assigned, depending on the background culture. This leads to a gap in recognition rates of expressions: this problem is present when displaying a robotic face too, as a robot's facial expression recognition is often hampered by a cultural divide, and poor scores of recognition rate may lead to poor acceptance and interaction. It would be desirable if robots could switch their output facial configuration flexibly, adapting to different cultural backgrounds. To achieve this, we made a generation system that produces facial expressions and applied it to the 24 degrees of freedom head of the humanoid social robot KOBIAN-R, and thanks to the work of illustrators and cartoonists, the system can generate two versions of the same expression, in order to be easily recognisable by both Japanese and Western subjects. As a tool for making recognition easier, the display of Japanese comic symbols on the robotic face has also been introduced and evaluated. In this work, we conducted a cross-cultural study aimed at assessing this gap in recognition and finding solutions for it. The investigation was extended to Egyptian subjects too, as a sample of another different culture. Results confirmed the differences in recognition rates, the effectiveness of customising expressions, and the usefulness of symbols display, thereby suggesting that this approach might be valuable for robots that in the future will interact in a multi-cultural environment.
机译:通过面部表情的情感显示是一个重要的沟通渠道。然而,在人类之间存在对面部线索的含义分配的差异,这取决于背景文化。这导致表达识别率的差距:当显示机器人面临时,这个问题也存在,因为机器人的面部表情识别经常被文化划分所阻碍,并且识别率的差的分数可能导致接受和相互作用差。如果机器人可以灵活地切换输出面部配置,适应不同的文化背景,是可取的。为实现这一目标,我们制作了一系列产生面部表情,并将其应用于人形社会机器人Kobian-R的24度,并且由于插画家和漫画人员的工作,系统可以生成两个版本的相同的表达,为了容易被日本和西方对象识别。作为使识别更容易的工具,还引入并评估了机器人面上的日本漫画符号的显示。在这项工作中,我们进行了一个跨文化研究,旨在评估这种差距的认可和寻找解决方案。该调查也延伸到埃及科目,作为另一种不同文化的样本。结果证实了识别率的差异,定制表达的有效性以及符号展示的有用性,从而表明这种方法对于未来将在多元文化环境中互动的机器人可能是有价值的。

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