See Hear! Multimodal Recording and Analysis Facility

见闻！

• 批准号: LE0668448
• 负责人: Prof Catherine Stevens
• 金额: $ 10万
• 依托单位: Western Sydney University
• 依托单位国家: 澳大利亚
• 项目类别: Linkage Infrastructure, Equipment and Facilities
• 财政年份: 2006
• 资助国家: 澳大利亚
• 起止时间: 2006-05-09 至 2007-12-30
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/LE0668448
• 关键词: See; Hear; Multimodal; Recording; Analysis

High resolution recording and analysis will exploit the full potential of motion capture with progress towards automatic recognition of gesture and, eventually, real-time systems. Automatic tracking and recognition systems are in high demand and the interlacing of data from multiple modes is now computationally achievable. SeeHear! will be coded using techniques in multimodal fusion - tracking of bodies will be enhanced by locating and recognizing facial features, and a learning algorithm used to classify gesture from patterns of force and physiological response. In the future, full interactivity will be achieved by interconnecting visual and auditory data with a flow on to applications in the performing arts, rehabilitation and security.

高分辨率的记录和分析将充分挖掘运动捕捉的潜力，朝着自动识别手势并最终实现实时系统的方向发展。自动跟踪和识别系统的需求很高，来自多种模式的数据的交错现在可以通过计算实现。看，听！将使用多模式融合技术进行编码-将通过定位和识别面部特征以及用于根据力和生理反应模式对手势进行分类的学习算法来增强对身体的跟踪。在未来，通过将视觉和听觉数据与流向表演艺术、康复和安全的应用程序相互连接，将实现完全的互动性。