Non-line-of-sight-imaging using dynamic relay surfaces

Area of Science:

• Optics
• Computational Imaging
• Photonics

Background:

• Non-line-of-sight (NLOS) imaging aims to visualize objects occluded from direct view.
• Current NLOS methods typically rely on static, flat relay surfaces for photon scattering.
• Dynamic environments pose a significant challenge for existing NLOS imaging techniques.

Purpose of the Study:

• To develop and validate a novel NLOS imaging approach capable of reconstructing scenes using non-planar and dynamic relay surfaces.
• To demonstrate that NLOS imaging can be performed effectively even when the relay surface changes during data acquisition.

Main Methods:

• Utilized a dual-detector system to capture time-of-flight (ToF) data from both the relay surface and the hidden scene.
• Developed a reconstruction algorithm that associates multiply-scattered photons with their origin on the relay surface.
• Enabled the system to compensate for the movement of the relay surface during data collection.

Main Results:

• Successfully performed NLOS imaging with a non-planar and rapidly changing relay surface.
• Reconstruction quality with a dynamic relay surface was comparable to that achieved with traditional static relay surfaces.
• The developed method accurately accounts for the changing positions of the relay surface.

Conclusions:

• This work overcomes a key limitation in NLOS imaging by enabling reconstructions in dynamic environments.
• The proposed technique significantly expands the applicability of NLOS imaging to real-world scenarios with moving surfaces.
• Future research can build upon this method for more robust imaging in complex, changing conditions.