About This Webinar
Establishing a standardized approach for evaluating Depth+RGB cameras is essential for objective comparisons and informed decision-making in robotics, smart manufacturing, and 3D scanning industry applications. This consistent evaluation framework aids in identifying the advantages and disadvantages of specific cameras under consideration, facilitating the selection of the optimal device for the target application.
Depth camera designs vary widely based on sensing technologies, optics, illumination, image processing, and other parameters. This session introduces a comprehensive testing methodology designed to evaluate Depth+RGB cameras using various accuracy and precision metrics, including depth accuracy, fill rate, spatial precision, and temporal precision. The testing process, data acquisition procedures, and data post-processing methods will be presented in detail. Additionally, comparative test results of selected commercial Depth+RGB cameras will be shared as a reference.
Attendees will gain a thorough understanding of the importance of a standardized evaluation framework for Depth+RGB cameras along with insights into a detailed testing methodology covering accuracy and precision metrics. Viewers will also learn different data acquisition and post-processing techniques and see comparative test results of popular Depth + RGB cameras, providing real-world examples for reference.
*** This presentation premiered during the
2024 Vision Spectra Conference. For more information on Photonics Media conferences and summits, visit
events.photonics.com
About the presenter
Xin Xie, Ph.D., received his doctorate in mechanical engineering from Oakland University, MI, and currently serves as the chief engineer and director of R&D at Orbbec, overseeing vision system design and hardware architecture. Previously, he was an assistant professor of mechanical engineering at Lawrence Technological University, MI. Xie’s publications in the field of optical measurement and sensing technologies include a scientific research book, 23 journal papers, and 14 conference papers with over 1,200 citations.