Aktuelles

Unlike traditional virtual reality head-mounted displays (HMDs), which confine users to a small tracking area and rely on external sensors, modern mixed reality (MR) HMDs can track user movements across larger spaces without external markers. This paper reviews several MR-HMDs currently on the market and evaluates the tracking accuracy and loop closure capabilities of four commercially available models across four distinct scenarios. Notably, HMDs with video see-through capabilities exhibit varying levels of visual degradation. Results from tracking experiments indicate that all devices perform consistently well in expansive, well-lit environments, with tracking accuracy remaining stable even in outdoor nighttime conditions. Additionally, most HMDs demonstrated effective error correction during loop closure, with errors in non-loop scenarios consistently exceeding those in loop scenarios.

Autor: Long Cheng

 

Download Whitepaper

          

Einführung

Virtual Reality (VR) is a computer-generated simulation of a fully immersive, three-dimensional environment that users experience through a head-mounted display (HMD) and controllers. Traditionally, VR has been defined as an experience that completely separates the user from the real world, making it necessary to track the user’s physical environment to manage the limited space. This separation leads to a conflict between the virtually unlimited space in VR and the finite physical space available in reality.

To mitigate potential safety risks from real-world objects, VR HMDs typically confine users to a defined tracking area, often smaller than 10×10 meters and shaped in a convex layout. Initially, tracking relied on external position sensors mounted on walls or ceilings, employing an "outside-in" tracking method. This approach used triangulation based on the distance of the user from these sensors, as illustrated in Figure 1(a). By the mid-2010s, newer VR HMDs incorporated black-and-white cameras, primarily for tracking space boundaries. An example is the Meta Quest 2, shown in Figure 1(b).

With advancements in computer vision and tracking algorithms, cameras gradually replaced traditional outside-in tracking devices. Research by Holzwarth et al. [1] demonstrated that consumer-grade camera-based tracking solutions can outperform external, outside-in tracking systems like SteamVR.

However, despite these technological advancements, VR experiences remain confined within limited physical tracking spaces due to the fully immersive nature of VR. Even the most accurate optical tracking solutions are constrained by the number of tracking stations available, making them unsuitable for large-scale applications.