Snapshot Interferometric 3D Imaging by Compressive Sensing and Deep Learning

Abstract

We demonstrate single-shot compressive three-dimensional (3D) $(x,y,z)$ imaging based on interference coding. The depth dimension of the object is encoded into the interferometric spectra of the light field, resulting a $(x,y,\lambda )$ datacube which is subsequently measured by a single-shot spectrometer. By implementing a compression ratio up to $400$, we are able to reconstruct $1G$ voxels from a 2D measurement. Both an optimization based compressive sensing algorithm and a deep learning network are developed for 3D reconstruction from a single 2D coded measurement. Due to the fast acquisition speed, our approach is able to capture volumetric activities at native camera frame rates, enabling 4D (volumetric-temporal) visualization of dynamic scenes.

16 pages, 12 figures

References in corpus (7)

• Res2Net: A New Multi-scale Backbone Architecture
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Coded aperture compressive temporal imaging
• Imaging With Nature: Compressive Imaging Using a Multiply Scattering Medium
• Low Photon Count Phase Retrieval Using Deep Learning
• One Network to Solve Them All --- Solving Linear Inverse Problems using Deep Projection Models
• Compressive Diffraction Tomography for Weakly Scattering