Fig. 14: Flowchart of the deep-learning-based fringe-enhancement method and the 3D reconstruction results of different approaches.

a The flowchart of the deep-learning-based fringe enhancement: the captured raw fringe images and the quality-enhanced versions are used to learn the mapping between the input fringe image and the output enhanced fringe part of the constructed DnCNN. b Input raw fringe pattern of a moving hand. c 3D reconstruction result obtained by traditional FT¹³⁸. d 3D reconstruction result obtained by the deep-learning method. a–d Adapted with permission from ref. ⁵¹, Optica Publishing