상세정보
학술대회
Accurate Fusing Multiple RGBD Sensors Using Geometry
- 발행일
- 201510
- 출처
- International Conference on Ubiquitous Robots and Ambient Intelligence (URAI) 2015, pp.18-21
- 협약과제
- 14PC6500, 병원의 광역 환경에 적용가능한 물류 로봇 시스템 개발, 이재연
- 초록
- This paper proposes an approach for fusing multiple RGBD sensors using geometry obtained from the captured image. We used Xtion Pro Live as the RGBD sensor. With the coupled line camera (CLC) method, the position of the camera is geometrically reconstructed when the size of a quadrilateral in the captured image is known. After roughly setting the hardware of multiple sensors on the frame, accurate calibration is proceeded by software using CLC. When positions of sensors are decided, the center of sensors can be derived. Input data of each sensors are reconstructed with this center point afterwards. Prior to the calibration, the method for connecting multiple sensors on a laptop is described and different square patterns used for calibration is discussed.
- KSP 제안 키워드
- Center point, Coupled line, RGB-D sensors, Xtion PRO, input data, multiple sensors, square pattern