الفهرس | Only 14 pages are availabe for public view |
Abstract In this thesis, we propose a complete 3D object detection and recognition system that is based on Microsoft Kinect sensor and contains an enhanced depth estimation algorithm. Our idea is to use a cheap on-the-shelf sensor and integrate/develop algorithms that use the sensors data to detect the objects{u2019} positions and define their type. An end-to-end Kinect based 3D object detection and recognition system was developed that has accurate results and high processing speed compared to literature |