videoseethrough

Videoseethrough is a display modality used in head-mounted displays in which the user sees the real environment through a live video feed captured by cameras, rather than through transparent optics. It contrasts with optical see-through, where the real world is seen directly through transparent lenses.

In a videoseethrough system, the cameras capture the scene and the resulting video is processed and composited with computer-generated imagery or annotations before being presented on the display. The display may be a single panel, a stereo pair, or a head-mounted screen. Accurate alignment, or registration, between the real-world video and virtual content depends on head tracking and calibration. Latency, resolution, and color fidelity are key technical challenges because delays or mismatches can cause discomfort and break perceived realism.

Applications include augmented reality and mixed reality work, industrial inspection, training, remote assistance, and telepresence, where reliable occlusion of virtual objects by real-world geometry is beneficial. Videoseethrough can provide robust depth cues and consistent lighting for overlays, but it can also suffer from video artifacts and limited exposure, which can affect perception in bright or dynamic environments.

Security and privacy concerns arise because cameras capture the surrounding environment. Users and bystanders may be recorded, and data protection practices are important. The choice between videoseethrough and optical see-through depends on use case, desired realism, and tolerance for latency.