Visual-based virtual reality and augmented reality have their own developments, but the fusion development trend of the two is unavoidable in the future. SLAM (simultaneous localization ad mapping) is a major component of the application of virtual reality and augmented reality, but there are still many challenging issues in terms of robustness. This paper proposes a vision-based virtual reality and augmented reality fusion technology, which can analyze and solve the problems such as device selection, tracking, motion interference, and plane recognition in the 3D scene reconstruction process. Finally, the challenging issues in SLAM are discussed.

virtual reality  /  augmented reality  /  SLAM  /  3D scene reconstruction