By integrating digital information with real-world scenes in real-time, augmented reality has become an important technology for enhancing viewing experience and understanding of the real world. At present, near-eye display (NED) is the commonly used tool to realize augmented reality owing to its portability and impact design. Generally, conventional NEDs commonly use lens to provide a virtual image at a fixed plane. However, this approach is often characterized by vergence-accommodation conflict, especially when displaying three-dimensional (3D) images. This undesirable phenomenon is mainly attributed to its inability to provide depth cues.
Although numerous approaches, such as tensor and holographic displays, have been developed to solve the VAC problem and provide depth cues, their display effect is still unsatisfactory. Recently, Maxwellian displays (also called retinal projection display) have drawn considerable research attention owing to their large field of view and high light efficiency. These displays use a narrow beam, allowing it to reduce the effective pupil size, resulting in a significant increase in the depth of the field (DOF). This feature also contributes to solving the VAC problem. Unfortunately, the effectiveness of Maxwellian displays is limited by their narrow eyebox.
Holographic Maxwellian displays based on wave optics have been identified as a promising approach for expanding the eyebox of Maxwellian displays. It allows flexible adjustment of DOF and viewpoint position. Like other Maxwellian displays, however, these holographic Maxwellian displays also fail to provide desirable depth cues for monocular vision. Therefore, developing effective approaches to overcome the above limitations of the previous holographic Maxwellian displays is highly desirable.
To overcome these limitations, Dr. Xu Zhang, Professor Guoqiang Lv and Professor Zi Wang from Hefei University of Technology proposed an innovative holographic super multi-view (SMV) Maxwellian NED. It was based on flexible wavefront modulation. Unlike previous methods, two or more parallax images of 3D objects captured from various viewpoints were prepared. Each image was multiplied with quadric phase distribution and converged to the corresponding viewpoints within the pupil to provide 3D vision. A series of simulations were performed to validate the applicability of the proposed method. Their work is currently published in the journal, Optics Letters.
The authors demonstrated the feasibility and effectiveness of the presented holographic SMV in solving the inherent challenges associated with previous Maxwellian displays, including providing depth cues for monocular vision. Moreover, its refresh rate was independent of the number of viewpoints. Despite the benefits of this approach, the coherence of the laser source induced serious cross talk between the viewpoints, which could potentially degrade its performance. Therefore, a time division method was proposed to avoid cross talks by destroying the temporal coherence. Furthermore, it provided a large DOF, flexible control of the viewpoint position and the ability to expand the horizontal eyebox through multiple spherical wave encoding.
In summary, this is the first study to report a holographic SMV Maxwellian display. Subsequent experiments demonstrated its ability to accurately reconstruct images at various depths without the effects of cross talk, while providing depth cues. Compared to the geometric optics based on SMV displays, the system proposed here was superior. It only required a single spatial light modulator without additional optical elements to achieve satisfactory compactness without lens aberration. In a statement to Advances in Engineering, Professor Zi Wang said the new method would contribute to designing highly efficient and multi-focal Maxwellian displays for application in augmented reality.
Zhang, X., Pang, Y., Chen, T., Tu, K., Feng, Q., Lv, G., & Wang, Z. (2022). Holographic Super multi-view Maxwellian near-eye display with eyebox expansion. Optics Letters, 47(10), 2530-2533.