Polarization 3D imaging technology: a review

Frontiers in Physics

Physics

Jiawei Song¹  Jinshou Wang¹  Cunying Pan¹  Zhiqiang Liu¹  Xuan Li²  Yudong Cai²  Xiaopeng Shao² 

[1] Hangzhou Institute of Technology, Hangzhou, China;Hangzhou Institute of Technology, Hangzhou, China;School of Optoelectronic Engineering, Xi’an, China;

关键词: polarization 3D imaging;    ambiguity normal;    single view;    passive imaging;    depth fusion;    deep learning;    challenge analysis;   

DOI  :  10.3389/fphy.2023.1198457

 received in 2023-04-01, accepted in 2023-04-18,  发布年份 2023

来源: Frontiers

【 摘 要 】

Polarization three-dimensional (3D) imaging technology has received extensive attention in recent years because of its advantages of high accuracy, long detection distance, simplicity, and low cost. The ambiguity in the normal obtained by the polarization characteristics of the target’s specular or diffuse reflected light limits the development of polarization 3D imaging technology. Over the past few decades, many shape from polarization techniques have been proposed to address the ambiguity issues, i.e., high-precision normal acquisition. Meanwhile, some polarization 3D imaging techniques attempt to extend experimental objects to complex specific targets and scenarios through a learning-based approach. Additionally, other problems and related solutions in polarization 3D imaging technology are also investigated. In this paper, the fundamental principles behind these technologies will be elucidated, experimental results will be presented to demonstrate the capabilities and limitations of these popular technologies, and finally, our perspectives on the remaining challenges of the polarization 3D imaging technology will be presented.

【 授权许可】

Unknown   
Copyright © 2023 Li, Liu, Cai, Pan, Song, Wang and Shao.