TY - JOUR
T1 - Deep-inverse correlography
T2 - Towards real-time high-resolution non-line-of-sight imaging
AU - Metzler, Christopher A.
AU - Heide, Felix
AU - Rangarajan, Prasana
AU - Balaji, Muralidhar Madabhushi
AU - Viswanath, Aparna
AU - Veeraraghavan, Ashok
AU - Baraniuk, Richard G.
N1 - Publisher Copyright:
© 2020 Optical Society of America.
PY - 2020
Y1 - 2020
N2 - Low signal-to-noise ratio (SNR) measurements, primarily due to the quartic attenuation of intensity with distance, are arguably the fundamental barrier to real-time, high-resolution, non-line-of-sight (NLoS) imaging at long standoffs. To better model, characterize, and exploit these low SNR measurements, we use spectral estimation theory to derive a noise model for NLoS correlography. We use this model to develop a speckle correlation-based technique for recovering occluded objects from indirect reflections. Then, using only synthetic data sampled from the proposed noise model, and without knowledge of the experimental scenes nor their geometry, we train a deep convolutional neural network to solve the noisy phase retrieval problem associated with correlography. We validate that the resulting deep-inverse correlography approach is exceptionally robust to noise, far exceeding the capabilities of existing NLoS systems both in terms of spatial resolution achieved and in terms of total capture time. We use the proposed technique to demonstrate NLoS imaging with 300 µm resolution at a 1 m standoff, using just two 1/8th s exposure-length images from a standard complementary metal oxide semiconductor detector.
AB - Low signal-to-noise ratio (SNR) measurements, primarily due to the quartic attenuation of intensity with distance, are arguably the fundamental barrier to real-time, high-resolution, non-line-of-sight (NLoS) imaging at long standoffs. To better model, characterize, and exploit these low SNR measurements, we use spectral estimation theory to derive a noise model for NLoS correlography. We use this model to develop a speckle correlation-based technique for recovering occluded objects from indirect reflections. Then, using only synthetic data sampled from the proposed noise model, and without knowledge of the experimental scenes nor their geometry, we train a deep convolutional neural network to solve the noisy phase retrieval problem associated with correlography. We validate that the resulting deep-inverse correlography approach is exceptionally robust to noise, far exceeding the capabilities of existing NLoS systems both in terms of spatial resolution achieved and in terms of total capture time. We use the proposed technique to demonstrate NLoS imaging with 300 µm resolution at a 1 m standoff, using just two 1/8th s exposure-length images from a standard complementary metal oxide semiconductor detector.
UR - http://www.scopus.com/inward/record.url?scp=85078196082&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85078196082&partnerID=8YFLogxK
U2 - 10.1364/OPTICA.374026
DO - 10.1364/OPTICA.374026
M3 - Article
AN - SCOPUS:85078196082
SN - 2334-2536
VL - 7
SP - 63
EP - 71
JO - Optica
JF - Optica
IS - 1
ER -