Object shapes in physical space are scanned and stored as point clouds for viewing in VR. Since point clouds are represented by huge numbers of points, the data volume becomes enormous, requiring study of appropriate sampling methods and compression coding techniques. When rendering point clouds, if the point cloud lacks attribute information, there is a problem where the backside or internal shapes of the point cloud cannot be seen. We propose the adaptive sampling method to solve this problem. We demonstrate that our proposed adaptive sampling is effective for human object recognition in general scenes containing multiple objects. To efficiently compress the adaptively sampled point cloud, we propose the layered representation extending deep learning-based compression methods. Experiments show that our proposed method can restore the adaptively sampled point cloud with high quality.

In this paper, the latest progress in high dynamic range (HDR) technologies for CMOS image sensors (CISs) is presented. In particular, three single-exposure HDR (SEHDR) technologies-multiple-conversion gain (CG), lateral overflow integration capacitor (LOFIC), and triple-quantization HDR (3Q-HDR)-developed by the author and his research organizations are summarized. In addition, recent progress and future prospects of CISs for sensing applications, enabled by advances in process, pixel, and circuit technologies, are discussed. This paper is an extended version of the previously published paper: “Progress in Development of Advanced High Dynamic Range CMOS Image Sensors,” by Ken Miyauchi in the Proceedings of the 32nd International Display Workshops (IDW '25).

Photochromic dye-doped liquid crystal devices change their properties in response to both optical and electrical stimuli, thereby enabling intelligent smart windows that can adapt to the environment. However, most devices investigated to date suffer from small absorbance that originates from the low solubility of dyes and thus requires a large cell-gap that hinders the electro-optic response. Here, we report the electro-optic properties of high concentration photochromic dye-doped nematic liquid crystals containing up to 30 wt% of diarylethene dye and their electro-optic and photochemical properties. In this system, a cell gap as thin as 2 μm was sufficient to achieve an absorbance exceeding 1, which facilitates a faster electro-optic response. We also observed a change in the response time in the activated and deactivated states, suggesting a change in the intrinsic material parameters by photoisomerization.

In Spatial Augmented Reality (SAR), conventional projection mapping often treats target objects as mere screens, potentially diminishing their inherent value. To address this, we propose “Material Transformation.” This approach changes the material appearance of an object while preserving its original physical structures, such as boundaries and joints.

To realize this, we generate textures using NoiseNCA. We evaluate two methods to preserve structural features: integrating edge images into texture generation and using region segmentation. Furthermore, reproducing texture gradients is essential for perceiving the 3D scene structure. We examine three methods for this: using illuminance gradients, area ratios, and a depth estimation model.

Quantitative evaluations show that the region segmentation method is superior for preserving structural features. For reproducing texture gradients, each method has specific suitable and unsuitable scenes. We clarify their effectiveness and limitations to realize Material Transformation effectively.

This paper proposes a dual-mode optical device utilizing electrically controlled gallium-based liquid metal (LM) droplets. By manipulating LM within an aqueous NaOH solution to maintain its metallic luster and high surface tension, the device operates in both reflective and transmissive modes. The former leverages the LM’s high reflectivity, while the latter employs its light-blocking capability as an optical shutter. These movements are driven by the Marangoni effect, induced by an electrically modulated interfacial tension gradient. We demonstrate a display application capable of maintaining images without a continuous power supply by integrating multiple miniaturized units. For actuation, an AC waveform with an 80% duty cycle enables smoother droplet control at lower voltages compared to conventional DC methods. This architecture offers a promising approach for low-power, multi-functional display technologies without mechanical moving parts.

We report a pilot study using an aerial display based on Aerial Imaging by Retro-Reflection (AIRR) to deliver barrier-free, life-sized visual stimuli to terrestrial animals. Predatory strike responses were observed in a Ghost mantis toward aerial images of crickets presented by the AIRR system. Although the number of observed responses was limited, these observations demonstrate the feasibility of using AIRR to present biologically relevant visual stimuli to terrestrial predators and highlight its potential as a platform for VR biology research.

Optical wireless power transmission (OWPT) requires converting incident light from multiple directions. Photovoltaic devices other than silicon solar cells are expensive, and the wavelength converted by the photovoltaic device must be close to the wavelength of the light source. We previously proposed a fixed concentrating receiver that concentrates incident light from multiple directions. Concentrating receiver allows reducing the area of the photovoltaic device, thereby lowering cost. The previously proposed receiver can concentrate light from multiple directions but suffers from light leakage issues. This study proposes adding liquid-crystal (LC) controlled lightguides to the previously proposed receiver to guide light to a stationary, smallarea power-generating element. Combining a high-refractive-index anisotropic LC with high-refractive-index glass (n=1.9) expands the range of incident angles for light switching to 19 deg. We successfully prototyped voltage-controlled light switching by a total reflection state.

This paper presents an analog AI accelerator utilizing current-programming memory based on Indium Gallium Zinc Oxide (IGZO) thin-film transistors (TFTs). Leveraging current-copier pixel circuits from organic light-emitting display (OLED) technology, the architecture performs high-precision multiply-accumulate (MAC) operations while serving as analog memory. IGZO’s ultra-low off-current facilitates long-term weight retention, significantly reducing power consumption by removing the need for frequent refreshes. To address analog non-idealities like feedthrough effects, an n-channel source follower is integrated, achieving output current deviations as low as 0.39% (corresponding to an equivalent precision of ∼8 bits over full scale). Furthermore, Monte Carlo simulations confirm robust parameter mismatch immunity, and system-level simulations on the MNIST dataset using the circuit-extracted error distribution validate minimal accuracy degradation (< 0.4%, achieving 97.4%). This design provides a scalable, energy-efficient solution for sustainable edge AI computing.