融合
结构光
计算机科学
计算机视觉
人工智能
光学
物理
语言学
哲学
作者
Peng Liu,Fan Xu,Jiaru Chu,Yuhang Chen
标识
DOI:10.1002/lpor.202402297
摘要
Abstract Structured light imaging, an advanced technology in computational 3D imaging, illuminates the target scene with structured light and then decodes the modulated light information to reconstruct the 3D object. The current structured light imaging systems are typically composed of a multitude of bulky diffractive optical elements and lenses, which impede their prospective integration and miniaturization. Furthermore, they are constrained to provide a narrow field of view and limited encoding information. Here, a multichannel fusion metasurface‐based structured light imaging method is proposed, which can project two sets of phase‐shifted fringes via a polarization multiplexed metasurface. This method allows for the encoding and analysis of more intricate information in the scene, including two sets of speckle patterns and two sets of fringe phases, by switching the orthogonal polarization orientation of the incident linearly polarized light. To fully exploit the modulated information, a speckle‐and‐phase fusion stereo reconstruction framework is developed, which enables high‐resolution and high‐precision 3D imaging via the fusion of multichannel information. This systematic solution for the metasurface‐based structured light imaging paves the way for its practical applications.
科研通智能强力驱动
Strongly Powered by AbleSci AI