Today, there is a growing interest in machines that operate and perform tasks in dynamic environment settings. These machines, most commonly found in robotic or autonomous navigation systems, pose many challenges that could be solved more effectively with knowledge of the environment the machine is residing in at the time. This context-awareness, or computer’s ability to understand the context of its environment plays an important role in many applications. The goal of the research refer here was to develop a solid framework for machines with computer vision to have an effective context-awareness ability. To this end, a range of methods and techniques were examined and a uniformed framework proposal were presented as a result. This framework strives for improved performances of machines in recognizing context components (objects, human and scenes) with the help of computer vision techniques. Experiments that were carried out throughout the research have not only demonstrated the framework’s effectiveness but also underlined positive impact it has on multiple aspects of computer vision.
|Date of Award||2016|
|Supervisor||Dat Tran (Supervisor), Wanli Ma (Supervisor) & Sharma D (Supervisor)|