This paper proposes a metasurface-enhanced structured light (SL) depth-sensing platform capable of scattering high-density (~10K) dot arrays over a 180° field of view (FOV). The platform utilizes subwavelength-scale light manipulation for enhanced FOV and diffraction efficiency compared to conventional methods. A proof-of-concept demonstration involving depth estimation of face masks is presented, utilizing a stereo matching algorithm. Furthermore, a high-throughput manufacturing method using nanoparticle-embedded resin (nano-PER) imprinting is demonstrated, enabling metasurface replication on various substrates. The authors suggest potential applications in face recognition and automotive robot vision.
Publisher
Nature Communications
Published On
Oct 10, 2022
Authors
Gyeongtae Kim, Yeseul Kim, Jooyeong Yun, Seong-Won Moon, Seokwoo Kim, Jaekyung Kim, Junkyeong Park, Trevon Badloe, Inki Kim, Junsuk Rho
Tags
metasurface
structured light
depth sensing
high-density dot arrays
face recognition
automotive vision
nanoparticle-embedded resin
Related Publications
Explore these studies to deepen your understanding of the subject.