Abstract
Camera calibration is a necessary preliminary step in computer vision for the estimation of the position of objects in the 3D world. Despite the intrinsic camera parameters can be easily computed offline, extrinsic parameters need to be computed each time a camera changes its position, thus not allowing for fast and dynamic network re-configuration. In this paper we present an unsupervised and automatic framework for the estimation of the extrinsic parameters of a camera network, which leverages on optimised 3D human mesh recovery from a single image, and which does not require the use of additional markers. We show how it is possible to retrieve the real-world position of the cameras in the network together with the floor plane, exploiting regular RGB images and with a weak prior knowledge of the internal parameters. Our framework can also work with a single camera and in real-time, allowing the user to add, re-position, or remove cameras from the network in a dynamic fashion.
Highlights
In computer vision and 3D reconstruction, many works over the years have tried to automate the process of camera resectioning and calibration
We propose our one-shot method for fully automatic and unsupervised camera network calibration that leverages on monocular 3D human pose estimation from single images
We presented a completely unsupervised and one-shot camera network calibration framework capable of calibrating a single camera or a camera network only from monocular human pose estimation cues
Summary
In computer vision and 3D reconstruction, many works over the years have tried to automate the process of camera resectioning and calibration. A recent trend in computer vision is pedestrian-based camera calibration, which focuses on finding how to estimate both intrinsic and extrinsic camera parameters by exploiting the cues provided by walking humans The main contribution, compared to the work in [7] consists of the capability of the system to obtain real-time camera network calibration, at comparable accuracy This is achieved thanks to the adoption of a faster SNWBP network [9] and a more precise human mesh recovery pipeline [16]. These improvements allow for an even easier deployment in real-world scenarios, and are helpful when dealing with large camera networks and real-time constraints
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
