- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Research on multi-camera calibration and point cloud correction method based on three-dimensional calibration object
摘要： The single camera measurement system can’t obtain all the surface information of the object limited by the field of view, thus it can’t achieve complete measurement of the object. Multi-camera system can overcome this difficulty. But it is difficult to unify the coordinate system of distinct cameras. In order to solve this problem, a global calibration method of multi-camera system is proposed. A three-dimensional cube calibration object is designed. Every camera is only calibrated on one surface of the cube calibration object. Because different surfaces of the cube calibration object are in a unified world coordinate, the coordinates of the feature points on different surfaces are naturally in the unified world coordinate system. Thus global calibration can be completed by calibrating multiple cameras at the same time. In addition, to minimize the error of calibration, a correction method is proposed. The correction parameters are obtained by the deviation between the world coordinates and the reconstructed coordinates of the feature points on calibration object. The parameters are applied to the measurement to improve the accuracy of the whole point cloud. Experiments are carried out on a multi-camera measurement system, and the results show that the method proposed in this paper is effective and feasible.
关键词： Calibration,Measurement,3D calibration object,Fringe projection,Multi-camera
更新于2025-09-23 15:23:52
-
[IEEE 2019 Photonics North (PN) - Quebec City, QC, Canada (2019.5.21-2019.5.23)] 2019 Photonics North (PN) - UV LEDs: Thermal Management, Applications, and Future Prospects
摘要： This paper proposes a generic methodology for the semi-automatic generation of reliable position annotations for evaluating multi-camera people-trackers on large video data sets. Most of the annotation data are automatically computed, by estimating a consensus tracking result from multiple existing trackers and people detectors and classifying it as either reliable or not. A small subset of the data, composed of tracks with insufficient reliability, is verified by a human using a simple binary decision task, a process faster than marking the correct person position. The proposed framework is generic and can handle additional trackers. We present results on a data set of ～6 h captured by 4 cameras, featuring a person in a holiday flat, performing activities such as walking, cooking, eating, cleaning, and watching TV. When aiming for a tracking accuracy of 60 cm, 80% of all video frames are automatically annotated. The annotations for the remaining 20% of the frames were added after human verification of an automatically selected subset of data. This involved ～2.4 h of manual labor. According to a subsequent comprehensive visual inspection to judge the annotation procedure, we found 99% of the automatically annotated frames to be correct. We provide guidelines on how to apply the proposed methodology to new data sets. We also provide an exploratory study for the multi-target case, applied on the existing and new benchmark video sequences.
关键词： Multi-camera tracking,people tracking,semi-automatic annotation,performance evaluation
更新于2025-09-19 17:13:59