Fast optimization of computer-generated holograms using conjugate gradient

Holographic display, as an ideal three-dimensional (3D) display method, can provide a realistic stereo-perception sense, and is expected to be used in the next generation of augmented reality (AR) prototypes without complex optical devices. While holographic 3D displays have been proposed in the past decades, high-quality holographic reconstructed images have always relied on well-designed generation algorithms, and it is still a challenge to obtain an accurate phase profile fast. In this work, we utilize a first-order optimization algorithm to optimize the phase profile using the conjugate gradient direction instead of the steepest gradient direction to update the phase parameters. We validate this proposed method by comparing the current widely used stochastic gradient method and the wirtinger holography method. The simulation and experimental results illustrate the superiority of the proposed method, which can achieve faster optimization with the same quality metrics.