Augmented Reality Tracking in Natural Environments

摘要: Tracking, or camera pose determination, is the main technical challenge in creating augmented realities. Constraining degree to which environment may be altered support tracking heightens challenge. This paper describes several years of work at USC Computer Graphics and Immersive Technologies (CGIT) laboratory develop self-contained, minimally intrusive systems for use both indoor outdoor settings. These hybrid-technology combine vision inertial sensing with research fiducial design, feature detection, motion estimation, recursive filters, pragmatic engineering satisfy realistic application requirements.