Deep Domain Adversarial Adaptation for Photon-efficient Imaging
This work addresses the domain shift issue in computational imaging for realistic photon-efficient applications, offering an incremental improvement by adapting models to real-world data without ground-truth labels.
The paper tackles the problem of domain shift in photon-efficient imaging, where models trained on simulated data perform poorly on real-world scenarios with different signal-to-background ratios and hardware properties. It presents a domain adversarial adaptation method that uses unlabeled real-world data, achieving superior performance in simulated and real-world experiments with a home-built imaging system.
Photon-efficient imaging with the single-photon light detection and ranging (LiDAR) captures the three-dimensional (3D) structure of a scene by only a few detected signal photons per pixel. However, the existing computational methods for photon-efficient imaging are pre-tuned on a restricted scenario or trained on simulated datasets. When applied to realistic scenarios whose signal-to-background ratios (SBR) and other hardware-specific properties differ from those of the original task, the model performance often significantly deteriorates. In this paper, we present a domain adversarial adaptation design to alleviate this domain shift problem by exploiting unlabeled real-world data, with significant resource savings. This method demonstrates superior performance on simulated and real-world experiments using our home-built up-conversion single-photon imaging system, which provides an efficient approach to bypass the lack of ground-truth depth information in implementing computational imaging algorithms for realistic applications.