Department of Bioengineering, University of California, Los Angeles, CA, USA.
Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, USA.
Nat Commun. 2021 Apr 12;12(1):2179. doi: 10.1038/s41467-021-22461-0.
Cameras with extreme speeds are enabling technologies in both fundamental and applied sciences. However, existing ultrafast cameras are incapable of coping with extended three-dimensional scenes and fall short for non-line-of-sight imaging, which requires a long sequence of time-resolved two-dimensional data. Current non-line-of-sight imagers, therefore, need to perform extensive scanning in the spatial and/or temporal dimension, restricting their use in imaging only static or slowly moving objects. To address these long-standing challenges, we present here ultrafast light field tomography (LIFT), a transient imaging strategy that offers a temporal sequence of over 1000 and enables highly efficient light field acquisition, allowing snapshot acquisition of the complete four-dimensional space and time. With LIFT, we demonstrated three-dimensional imaging of light in flight phenomena with a <10 picoseconds resolution and non-line-of-sight imaging at a 30 Hz video-rate. Furthermore, we showed how LIFT can benefit from deep learning for an improved and accelerated image formation. LIFT may facilitate broad adoption of time-resolved methods in various disciplines.
高速摄像机正在为基础科学和应用科学领域的技术提供支持。然而,现有的超高速摄像机无法应对扩展的三维场景,也无法实现非视距成像,而非视距成像需要一长串时间分辨的二维数据。因此,当前的非视距成像仪需要在空间和/或时间维度上进行广泛的扫描,这限制了它们只能用于静态或缓慢移动的物体成像。为了解决这些长期存在的挑战,我们提出了超快光场层析成像(LIFT),这是一种瞬态成像策略,提供了超过 1000 的时间序列,并能够实现高效的光场采集,从而实现完整的四维空间和时间的快照采集。利用 LIFT,我们展示了飞行现象的三维成像,分辨率达到<10 皮秒,非视距成像的视频帧率达到 30 Hz。此外,我们还展示了 LIFT 如何受益于深度学习以实现改进和加速的图像形成。LIFT 可能会促进时间分辨方法在各个领域的广泛应用。