LiDAR Point Cloud Virtual Reality Visual Evaluation Through Structural Similarity Index Measure and Edge Metric Similarity

VR has become one of the most reliable simulation techniques due to the practicality of human-subject research. Meanwhile, VR simulation technology aims to achieve the best photorealistic environment for the best immersive experience. However, wayfinding VR simulation with a point cloud environment must be evaluated to determine its feasibility in identifying wayfinding variables. Therefore, this study will analyse the visual quality of point cloud presentation in VR using near and far point observation, SSIM (Structural Similarity Index Measure), edge similarity, Frame Per Second (FPS), and flicker visibility. Various VR point cloud densities were assessed against ground truth images. The result indicated that the higher the density, the better the visual quality for the optimum VR simulation. Nonetheless, a 10 million point cloud density is sufficient for VR simulation, even though 20 million provides a denser and more detailed point representation