Exploiting Loops in the Camera Array for Automatic Focusing Depth Estimation

Abstract Autofocus is a fundamental and key problem for modern imaging sensor design. Although this problem has been well studied in single camera literature, unfortunately, little research has been done on large-scale camera arrays. Most of the existing synthetic aperture imaging systems still need to manually select the optimal focus plane when an object moves. Unlike the conventional autofocus method, which sweeps the focus plane to find the maximal contrast, we present a novel optimization framework to handle the above challenges. In particular, we formulate the camera array autofocus problem as a constrained optimization problem by minimizing the temporal and spatial correspondences error subject to global loop constraint. Then this problem is relaxed as a quadratic program and solved using sequential quadratic programming. The experimental results show that the proposed method achieves a better performance compared with the results of traditional methods. To the best of our knowledge, our proposed method is the first optimization framework for solving camera array autofocus problem and it is of great importance to improve the performance of the existing synthetic aperture imaging system.