Hot Paper - Deep-learning-enabled dual-frequency composite fringe projection profilometry for single-shot absolute 3D shape measurement

Deep-learning-enabled dual-frequency composite fringe projection profilometry for single-shot absolute 3D shape measurement

支持深度学习的双频复合条纹投影轮廓测量，用于单次绝对 3D 形状测量
シングルショット絶対3D形状測定のための深層学習対応のデュアル周波数複合フリンジプロジェクションプロフィロメトリー
단일 샷 절대 3D 형상 측정을 위한 딥 러닝 지원 이중 주파수 합성 프린지 투영 프로파일로메트리
Perfilometría de proyección de franjas compuesta de frecuencia dual habilitada para aprendizaje profundo para la medición de forma 3D absoluta de disparo único
Profilométrie de projection de franges composite à double fréquence activée par apprentissage en profondeur pour une mesure de forme 3D absolue en un seul coup
Двухчастотная составная интерференционная профилометрия с возможностью глубокого обучения для однократного измерения абсолютной трехмерной формы

Yixuan Li 李奕萱 ¹ ², Jiaming Qian 钱佳铭 ¹ ², Shijie Feng 冯世杰 ¹ ², Qian Chen 陈钱 ¹ ², Chao Zuo 左超 ¹ ²

¹ Smart Computational Imaging (SCI) Laboratory, Nanjing University of Science and Technology, Nanjing 210094, China
中国南京南京理工大学智能计算成像实验室
² Jiangsu Key Laboratory of Spectral Imaging & Intelligent Sense, Nanjing University of Science and Technology, Nanjing 210094, China
中国南京南京理工大学江苏省光谱成像与智能感知重点实验室

https://doi.org/10.29026/oea.2022.210021

Opto-Electronic Advances, 10 March 2022

Abstract

Single-shot high-speed 3D imaging is important for reconstructions of dynamic objects. For fringe projection profilometry (FPP), however, it is still challenging to recover accurate 3D shapes of isolated objects by a single fringe image.

In this paper, we demonstrate that the deep neural networks can be trained to directly recover the absolute phase from a unique fringe image that involves spatially multiplexed fringe patterns of different frequencies. The extracted phase is free from spectrum-aliasing problem which is hard to avoid for traditional spatial-multiplexing methods.

Experiments on both static and dynamic scenes show that the proposed approach is robust to object motion and can obtain high-quality 3D reconstructions of isolated objects within a single fringe image.

Reviews and Discussions