Computer vision for enhanced communications

增强通信的计算机视觉：互联网、数据库管理、广播、移动电话、多传感器系统、机器人和监控等应用中的多媒体通信系统，不断产生、传输和接收大量可视数据。除非处理系统能够有效地从这些数据中提取有用的信息，否则应用程序的性能将很差，从而导致服务质量下降和失去经济机会。在这种情况下，通信内容分析和解释对于富有成效的应用程序至关重要。这项视觉通信研究是INRS-EMT多媒体研究计划的一部分。其目的是开发高效的计算机视觉方法，以加快和增强视觉通信。它研究了视觉数据分析和解释中的两个广泛的主题：图像分割和图像序列中的运动分析。这些主题关键问题的预期解决方案有望实现丰富、高效的新图像判读技术。这些技术将服务于医疗诊断和干预、机器人和安全等关键应用中的可视通信。

COMPUTER VISION FOR ENHANCED COMMUNICATIONS: Multimedia communication systems in applications such as internet, database management, broadcasting, mobile telephony, multisensor systems, robotics, and surveillance, continually produce, transmit, and receive massive amounts of visual data. Unless processing systems are effective at extracting useful information from this data, applications will perform poorly, resulting in degraded quality of service and loss of economic opportunity. In this context, communication content analysis and interpretation are vital to productive applications. This research in visual communications is part of the multimedia research programme at INRS-EMT. Its purpose is to develop efficient computer vision methods to expedite and enhance visual communications. It investigates two broad subjects in analysis and interpretation of visual data: image segmentation and motion analysis in image sequences. The anticipated solution of these subjects key problems promises to enable a rich array of efficient new image interpretation techniques. These techniques will serve visual communications in critical such applications as medical diagnostic and intervention, robotics, and security.