Abstract: As a key technology of the integration of humans, machines, and objects, spatial sensing technology supports in-depth interactions across physical, informational, and social spaces through the provision of spatial information. Currently, this technology is undergoing a significant transformation from the macro world to the micro world, with its sensing precision advancing from the decimeter/centimeter level to the millimeter/sub-millimeter level. However, traditional computer vision-driven spatial sensing techniques are constrained by the limited spatial resolution of cameras, making it challenging to achieve super-resolution sensing in the three-dimensional micro world. As a novel visual sensing approach, Moiré patterns offer a fresh perspective for tackling ultra-high-precision spatial sensing challenges due to their low-frequency amplification effect and robust frequency domain characteristics. This article systematically reviews the latest research advancements in ultra-high-precision spatial sensing based on Moiré pattern vision, focusing on three key aspects: constructing sensing models, enhancing spatial generalization capabilities, and improving temporal generalization capabilities.