Inward three-dimensional scene acquisition virtual compound eye camera

Abstract

The invention relates to the field of digital three-dimensional scene construction, and provides an inward three-dimensional scene acquisition virtual compound eye camera. The camera comprises a planning module for realizing camera lens layout planning based on an enveloping sphere scheme; an acquisition module which is formed by integrating a camera lens for acquiring images, a camera lens pose gyroscope, a GPS positioner, a UWB positioner and a wireless transmitter; a control module which is composed of a network control server; a storage module which is composed of an image acquisition cardand a data storage card. Through the Internet of Things, the functional modules form a dynamic scene acquisition system with layout coordination, information coordination, action coordination, function coordination and task coordination. According to the invention, the acquired scene information and the manufactured three-dimensional scene have space-time consistency, real-time dynamics, pose sensibility and random integration; the acquisition information is not influenced by the scene scale, and no matter how large the scene scale is, dynamic reconstruction of the camera scale, the acquisition orientation and the like can be realized through enveloping sphere planning.

Description

technical field [0001] The invention relates to the field of digital three-dimensional scene construction, in particular to an introverted three-dimensional scene acquisition virtual compound eye camera. Background technique [0002] The essence of dynamic digital 3D scene construction is to meet the requirements of "four characteristics" such as "time-space consistency, real-time dynamics, pose sensitivity, and random integration" to meet the requirements of "heaven and earth position, topography, spatial distribution, and geometric size." Three-dimensional, visualized, and immersive VR / AR / MR / XR scenes with one-to-one correspondence such as "object-position-time" required by the "four shapes". [0003] 1. To construct a dynamic digital 3D scene with "four characteristics" and "four shapes", the basic premise is to implement synchronous acquisition of all-round and full-point acquisition under the control of a unified clock and with the cross-section at the same time as the ...

AI Technical Summary

Problems solved by technology

[0009] 1. Regardless of whether it is a long object or a round object, if you do not move the camera position or rotate the camera angle, you cannot satisfy the imaging field of view of 360° horizontally + 360° vertically

[0010] 2. Rotate the camera angle, move the tripod, move on the track, etc. There is a certain time interval for the collection of each photo, and the photos taken do not meet the time-space consistency requirements

[0011] 3. Regardless of whether it is a long object or a round object, it can only be spliced ​​to construct a horizontal 360° + vertical 360° static scene, but not a dynamic scene

[0012] 4. When the scene is large, it is difficult to accurately determine the collection point required by the above scheme, and even when the collection point can be determined, the photographer and the lens cannot reach the collection point

Images