High-quality reconstruction in non-line-of-sight imaging via transient multiplied photon count data

Imaging scenes or detecting objects hidden from a camera's view have extensive applications. Previous studies have recently demonstrated impressive results in time-of-flight-based non-line-of-sight (NLOS) imaging systems, but several limitations remain. Measurements are captured by time-resolved detectors based on scanning sampling spots at the diffuse wall. Setting numerous sampling points slows down the process of data acquisition, but reconstruction quality cannot be ensured with limited sampling points. Here, we propose a method, transient interpolation to original sample data, to solve the contradiction between the time of data collection and imaging quality. We calculate the flight time of photons and simulate the photon count data with 15×15 sampling points and multiply them based on our method to obtain high-quality reconstruction within a shorter time. The simulated experiment shows that compared to the original data, transient multiplied data achieves high-quality reconstruction and shortens the measurements-captured time theoretically. Furthermore, our method is robust to different algorithms and can potentially be used in real-time imaging fields.