Stochastic Light Field Holography on Experimental Hardware. Recent advances in holographic displays achieve high image fidelity using smooth-phase holograms. This comes at the cost of a highly concentrated eye box limiting image formation to only the central view. For a more robust visual experience, holographic displays should provide good image quality for arbitrary pupil-states (diameter, location, and accommodation) within the entire available eye box. By definition requires a random-phase hologram. With Stochastic Light Field Holography (SLFH), we propose a novel framework that ensures photo-consistency over the entire eye-box volume. We implement a novel Focal Stack supervision algorithm (LF2FS) and show that SLFH is a generalization of state-of-the-art (SOTA) (Focal Stack and Short-Time-Fourier-Transform (STFT)) supervision of CGH optimization algorithms, representing only a limited subset of possible pupil states. Consequently, our proposed method provides the best average image fidelity over the entire eyebox compared to SOTA. On the left, we show experimental captures from three example pupil states (Front/Back focus for accommodation, left/right for parallax). On the right, we show epipolar slices generated by capturing a 1D trajectory consisting of 31 small pupils (1/8th of eyebox) from left to right for both front and back focus. Our SLFH method produces the least artifacts over the full eyebox, while Focal Stack supervision produces strong color-fringing at occlusions and STFT-optimized holograms often show ringing artifacts and over-sharpened in focus images.

Abstract: The Visual Turing Test is the ultimate goal to evaluate the realism of holographic displays. Previous studies have focused on addressing challenges such as limited étendue and image quality over a large focal volume, but they have not investigated the effect of pupil sampling on the viewing experience in full 3D holograms. In this work, we tackle this problem with a novel hologram generation algorithm motivated by matching the projection operators of incoherent (Light Field) and coherent (Wigner Function) light transport. To this end, we supervise hologram computation using synthesized photographs, which are rendered on-the-fly using Light Field refocusing from stochastically sampled pupil states during optimization. The proposed method produces holograms with correct parallax and focus cues, which are important for passing the Visual Turing Test. We validate that our approach compares favorably to state-of-the-art CGH algorithms that use Light Field and Focal Stack supervision. Our experiments demonstrate that our algorithm improves the viewing experience when evaluated under a large variety of different pupil states.