iToF Depth

Indirect Time-of-Flight (iToF) depth sensing offers a cost-effective method for capturing dense 3D information, but suffers from inaccuracies caused by multipath interference (MPI) and motion artifacts. Current research focuses on developing computational models and deep learning architectures to mitigate these issues, employing techniques like polarimetric imaging and weakly-supervised optical flow estimation to improve depth accuracy and robustness. These advancements are significant for improving the reliability and applicability of iToF technology in various fields, including robotics, autonomous driving, and augmented reality.