The current measurement methods of strabismus include the corneal light reflection method, prism alternate covering, etc., which especially rely on the subjective experience of doctors, and there is a large error between different measurers， leading to serious underestimation of strabismus prevalence and insufficient care for strabismus patients. Here, the investigators established and validated an artificial intelligence system to achieve an automatic diagnosis of strabismus based on patient-sourced videos of programmatic cover tests. Three-dimensional reconstruction methods are used to digitize the parameters of head and eye positions. This system has been integrated into a smartphone platform to be further validated through hospital-based and population-based clinical trials.