Fast AI Autofocus for Near-Eye Display Testing: A Solution for Shallow Depth of Field and Large Focusing Range

Abstract

In recent years, the rapid development of Extended Reality (XR) technologies, including Augmented Reality (AR) and Virtual Reality (VR), has led to an increasing demand for high-precision testing of near-eye display (NED) devices. These devices require accurate autofocus mechanisms due to challenging optical conditions such as small depth-of-field (DoF) and large focus travel. Traditional autofocus methods, reliant on manual experience or contrast-based algorithms, are inefficient and unstable under complex lighting conditions. This paper presents an innovative AI-assisted autofocus method designed to optimize traditional autofocus algorithms, integrating deep learning models with iterative optimization techniques to enhance focus precision and speed. The proposed method achieves rapid autofocus, overcoming the shortcomings of conventional techniques, while ensuring high reliability and consistency in challenging test scenarios.