Real-time gesture recognition of unmanned aerial vehicle video

无人机视频实时手势识别

• 批准号: 486479-2015
• 负责人: Androutsos, Dimitrios
• 金额: $ 1.82万
• 依托单位: Ryerson University
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=585629
• 关键词: Real; time; gesture; recognition; unmanned

This proposal concentrates on the research, design and optimization of a real-time video gesture recognition system that can be applied to the video feed coming from a camera mounted on a drone (or Unmanned Aerial Vehicle). The system requires novel algorithm to recognize motions in sequential video frames with a complex, moving background. It will predict and analyze the global motion of the camera and the local motion of the objects simultaneously based on probabilistic graphic model and statistical methods. For example, one main goal of the project is to be able to identify in near real-time when people in a crowd wave up at the drone. Detection of such an event would then trigger a photograph and integration into social media.

本课题主要研究、设计和优化一种实时视频手势识别系统，该系统可以应用于安装在无人机（或无人驾驶飞行器）上的摄像机的视频馈送。该系统需要新颖的算法来识别具有复杂运动背景的连续视频帧中的运动。它将基于概率图形模型和统计方法同时预测和分析摄像机的全局运动和物体的局部运动。例如，该项目的一个主要目标是能够在接近实时的情况下识别人群中有人向无人机挥手。检测到这样的事件，就会触发一张照片，并融入社交媒体。