This repository contains methods and selected results related to the analysis of tracking and SLAM reliability in standalone VR systems (e.g. Meta Quest).
The focus is on time-based analysis of tracking stability, combining runtime tracking logs with auxiliary signals (e.g. video-based metrics or annotations) to identify degradation and failure events.
It also includes statistical analysis methods used to evaluate reliability trends and compare experimental conditions.
The methodology is based on:
Time synchronization of heterogeneous data sources
- Window-based analysis of tracking behavior
- Use of robust, data-driven baselines
- Classification of tracking reliability into discrete states
The repository also includes partial results and visualizations derived from the associated research article.
This repository is intended as:
- A methodological reference
- A reproducible complement to the publication
- A starting point for further analysis of VR tracking reliability