A study on recognizing multi-real world object and estimating 3D position in augmented reality

Abstract As augmented reality technologies develop, real-time interactions between objects present in the real world and virtual space are required. Generally, recognition location estimation carried out using tracking techniques, typically markers. However, markers creates spatial constraints simultaneous of objects. Therefore, we propose a system that enables camera visualizes visual information through positioning We scanned an RGB-D camera. A three-dimensional (3D) dense point cloud map is created clouds generated video images. Among information, detected retrieved based on pre-learned data. Finally, predicted pose objects, other may be augmented. Our estimates object 3D simple enabling viewing location.