This work introduces a perspective-corrected video see-through mixed-reality head-mounted display with edge-preserving occlusion and low-latency capabilities. To realize the consistent spatial and temporal composition of a captured real world containing virtual objects, we perform three essential tasks: 1) to reconstruct captured images so as to match the user's view; 2) to occlude virtual objects with nearer real objects, to provide users with correct depth cues; and 3) to reproject the virtual and captured scenes to be matched and to keep up with users' head motions. Captured image reconstruction and occlusion-mask generation require dense and accurate depth maps. However, estimating these maps is computationally difficult, which results in longer latencies. To obtain an acceptable balance between spatial consistency and low latency, we rapidly generated depth maps by focusing on edge smoothness and disocclusion (instead of fully accurate maps), to shorten the processing time. Our algorithm refines edges via a hybrid method involving infrared masks and color-guided filters, and it fills disocclusions using temporally cached depth maps. Our system combines these algorithms in a two-phase temporal warping architecture based upon synchronized camera pairs and displays. The first phase of warping is to reduce registration errors between the virtual and captured scenes. The second is to present virtual and captured scenes that correspond with the user's head motion. We implemented these methods on our wearable prototype and performed end-to-end measurements of its accuracy and latency. We achieved an acceptable latency due to head motion (less than 4 ms) and spatial accuracy (less than 0.1° in size and less than 0.3° in position) in our test environment. We anticipate that this work will help improve the realism of mixed reality systems.

We propose a varifocal occlusion technique for optical see-through head-mounted displays (OST-HMDs). Occlusion in OST-HMDs is a powerful visual cue that enables depth perception in augmented reality (AR). Without occlusion, virtual objects rendered by an OST-HMD appear semi-transparent and less realistic. A common occlusion technique is to use spatial light modulators (SLMs) to block incoming light rays at each pixel on the SLM selectively. However, most of the existing methods create an occlusion mask only at a single, fixed depth-typically at infinity. With recent advances in varifocal OST-HMDs, such traditional fixed-focus occlusion causes a mismatch in depth between the occlusion mask plane and the virtual object to be occluded, leading to an uncomfortable user experience with blurred occlusion masks. In this paper, we thus propose an OST-HMD system with varifocal occlusion capability: we physically slide a transmissive liquid crystal display (LCD) to optically shift the occlusion plane along the optical path so that the mask appears sharp and aligns to a virtual image at a given depth. Our solution has several benefits over existing varifocal occlusion methods: it is computationally less demanding and, more importantly, it is optically consistent, i.e., when a user loses focus on the corresponding virtual image, the mask again gets blurred consistently as the virtual image does. In the experiment, we build a proof-of-concept varifocal occlusion system implemented with a custom retinal projection display and demonstrate that the system can shift the occlusion plane to depths ranging from 25 cm to infinity.

Spatial augmented reality (SAR) pursues realism in rendering materials and objects. To advance this goal, we propose a hybrid SAR (HySAR) that combines a projector with optical see-through head-mounted displays (OST-HMD). In an ordinary SAR scenario with co-located viewers, the viewers perceive the same virtual material on physical surfaces. In general, the material consists of two components: a view-independent (VI) component such as diffuse reflection, and a view-dependent (VD) component such as specular reflection. The VI component is static over viewpoints, whereas the VD should change for each viewpoint even if a projector can simulate only one viewpoint at one time. In HySAR, a projector only renders the static VI components. In addition, the OST-HMD renders the dynamic VD components according to the viewer's current viewpoint. Unlike conventional SAR, the HySAR concept theoretically allows an unlimited number of co-located viewers to see the correct material over different viewpoints. Furthermore, the combination enhances the total dynamic range, the maximum intensity, and the resolution of perceived materials. With proof-of-concept systems, we demonstrate HySAR both qualitatively and quantitatively with real objects. First, we demonstrate HySAR by rendering synthetic material properties on a real object from different viewpoints. Our quantitative evaluation shows that our system increases the dynamic range by 2.24 times and the maximum intensity by 2.12 times compared to an ordinary SAR system. Second, we replicate the material properties of a real object by SAR and HySAR, and show that HySAR outperforms SAR in rendering VD specular components.

We propose a hybrid SAR concept combining a projector and Optical See-Through Head-Mounted Displays (OST-HMD). Our proposed hybrid SAR system utilizes OST-HMD as an extra rendering layer to render a view-dependent property in OST-HMDs according to the viewer's viewpoint. Combined with view-independent components created by a static projector, the viewer can see richer material contents. Unlike conventional SAR systems, our system theoretically allows unlimited number of viewers seeing enhanced contents in the same space while keeping the existing SAR experiences. Furthermore, the system enhances the total dynamic range, the maximum intensity, and the resolution of perceived materials. With a proof-of-concept system that consists of a projector and an OST-HMD, we qualitatively demonstrate that our system successfully creates hybrid rendering on a hemisphere object from five horizontal viewpoints. Our quantitative evaluation also shows that our system increases the dynamic range by 2.1 times and the maximum intensity by 1.9 times compared to an ordinary SAR system.

We propose an occlusion compensation method for optical see-through head-mounted displays (OST-HMDs) equipped with a singlelayer transmissive spatial light modulator (SLM), in particular, a liquid crystal display (LCD). Occlusion is an important depth cue for 3D perception, yet realizing it on OST-HMDs is particularly difficult due to the displays' semitransparent nature. A key component for the occlusion support is the SLM-a device that can selectively interfere with light rays passing through it. For example, an LCD is a transmissive SLM that can block or pass incoming light rays by turning pixels black or transparent. A straightforward solution places an LCD in front of an OST-HMD and drives the LCD to block light rays that could pass through rendered virtual objects at the viewpoint. This simple approach is, however, defective due to the depth mismatch between the LCD panel and the virtual objects, leading to blurred occlusion. This led existing OST-HMDs to employ dedicated hardware such as focus optics and multi-stacked SLMs. Contrary to these viable, yet complex and/or computationally expensive solutions, we return to the single-layer LCD approach for the hardware simplicity while maintaining fine occlusion-we compensate for a degraded occlusion area by overlaying a compensation image. We compute the image based on the HMD parameters and the background scene captured by a scene camera. The evaluation demonstrates that the proposed method reduced the occlusion leak error by 61.4% and the occlusion error by 85.7%.