Mobile Augmented Reality System For Smart Phones

Mobile augmented reality, which is augmented reality in essence and applied to the mobile devices, has become one of the hot spots of the field of human-computer interaction. Because of the lack of computing ability and special hardware architecture, the tracking and registration technology and interaction used on PC is no more suitable for mobile device. Motived by this problem, we provide a multi-scene effective mobile augmented reality system, which is applied to smart phone. This system combine the non-visual sensor based context-aware and computer vision based tracking and registration technology, thus enabling it to enhance user’s perception of the surrounding environment in different scenes. In outdoors, non-visual sensors are used to track POI in long distance and the user gets a general understanding about the POI, such as name, orientation and distance from the uses themselves. While, natural feature based tracking is used to recognize and track the objects indoors. The system provides multi-media resource to the user. Besides, we focus on augmented information overlaying on the size limited screen and also the usability of this mobile augmented reality system. Our key contributions in this thesis are as follows:1) Research a mobile augmented reality system to improve the user’s awareness of the real environment in multi-scenes.2) Combine non-visual sensor based context-aware and computer vision based tracking to improve the accuracy of the tracking and registration.3) Realize real-time natural feature based tracking and registration on the mobile platform with limited computational performance and memory space.