Table of Contents
Fetching ...

Head-wearable Holographic Head-mounted Display with 6 Degrees of Freedom

Taichi Sakakihara, Teppei Jodo, Seok Kang, Yuji Sakamoto

TL;DR

The paper tackles the challenge of real-time hologram data generation for holo-HMDs with six degrees of freedom, which is essential for head-motion–synchronized 3D views and reduced VAC. It introduces a fast phase-correction CGH algorithm that precomputes object-light and applies motion-dependent phase corrections, achieving CGH in $O(P)$ time independent of the number of object sources $N$. A palm-sized holo-HMD prototype with a concave-mirror optical path, full-color TD multiplexing, and a head-motion sensor demonstrates real-time rendering at approximately 40 fps and a theoretical FOV near $18^{\circ}$. The results show depth expressivity and robust six-DoF reconstruction, indicating practical viability while acknowledging challenges in FOV, speckle, and assembly sensitivity. This work provides a practical blueprint for developing commercially viable holo-HMDs and invites further optimization of optical design and speckle handling.

Abstract

A head-mounted display (HMD) using holography technology (holo-HMD) is expected to be the next generation of HMDs capable of reducing three-dimensional sickness. In HMDs, it is important to generate images that respond to head movement in real time. However, in holo-HMDs, generation of hologram data in real time is difficult due to the large computational resources required. This paper proposes a fast calculation algorithm for generating hologram data for holo-HMDs, which requires low computational power. A holo-HMD supporting six degrees of freedom was also developed using this algorithm and it was confirmed that it obtained reconstructed images with six degrees of freedom in real time (30 fps or more).

Head-wearable Holographic Head-mounted Display with 6 Degrees of Freedom

TL;DR

The paper tackles the challenge of real-time hologram data generation for holo-HMDs with six degrees of freedom, which is essential for head-motion–synchronized 3D views and reduced VAC. It introduces a fast phase-correction CGH algorithm that precomputes object-light and applies motion-dependent phase corrections, achieving CGH in time independent of the number of object sources . A palm-sized holo-HMD prototype with a concave-mirror optical path, full-color TD multiplexing, and a head-motion sensor demonstrates real-time rendering at approximately 40 fps and a theoretical FOV near . The results show depth expressivity and robust six-DoF reconstruction, indicating practical viability while acknowledging challenges in FOV, speckle, and assembly sensitivity. This work provides a practical blueprint for developing commercially viable holo-HMDs and invites further optimization of optical design and speckle handling.

Abstract

A head-mounted display (HMD) using holography technology (holo-HMD) is expected to be the next generation of HMDs capable of reducing three-dimensional sickness. In HMDs, it is important to generate images that respond to head movement in real time. However, in holo-HMDs, generation of hologram data in real time is difficult due to the large computational resources required. This paper proposes a fast calculation algorithm for generating hologram data for holo-HMDs, which requires low computational power. A holo-HMD supporting six degrees of freedom was also developed using this algorithm and it was confirmed that it obtained reconstructed images with six degrees of freedom in real time (30 fps or more).
Paper Structure (27 sections, 27 equations, 22 figures, 2 tables)

This paper contains 27 sections, 27 equations, 22 figures, 2 tables.

Figures (22)

  • Figure 1: point-light method. An object is represented as a collection of point light sources, and object light on hologram plane is obtained as synthesis of spherical waves from each point light source.
  • Figure 2: Schematic diagram of holometric video streaming (HVS) for broadcasting.
  • Figure 3: Schematic of optical system of our holo-HMD prototype. Laser light is collimated using convex lens and modulated using hologram displayed on SLM to form real image, and image is magnified using a concave mirror.
  • Figure 4: Magnification using concave mirror. Light wavefront modulated by the SLM forms real image $P_r(y_r,z_r)$, which is magnified using concave mirror, such as magnifying glass, and magnified virtual image
  • Figure 5: Schematic of optical system and theoretical maximum FOV of our holo-HMD prototype. An SLM generates light that forms a reconstructed image, which is magnified the a convex lens.
  • ...and 17 more figures