A Subjective Quality Evaluation of 3D Mesh with Dynamic Level of Detail in Virtual Reality
Duc Nguyen, Tran Thuy Hien, Truong Thu Huong
TL;DR
This work addresses the problem of delivering high-quality VR experiences under resource constraints by dynamically adjusting the Level of Detail (LoD) of 3D meshes. It adopts a subjective DSIS study in a VR setup with five meshes, four LoD levels, and four viewing distances to quantify perceived quality, revealing distance- and content-dependent effects on MOS. A distance- and geometry-aware MOS model is proposed, using the relation $MOS = \alpha\ln(Q) + \beta\ln(D) + \gamma\ln(Q)\ln(D) + \delta$, and is shown to achieve high predictive accuracy (e.g., $R^2 \ge 0.97$, $PLCC,SROCC > 0.98$, $RMSE < 0.10$). The findings indicate substantial tolerance to vertex reduction for large meshes and provide a practical QoE predictor to guide LoD selection in VR systems.
Abstract
3D meshes are one of the main components of Virtual Reality applications. However, many network and computational resources are required to process 3D meshes in real-time. A potential solution to this challenge is to dynamically adapt the Level of Detail (LoD) of a 3D mesh based on the object's position and the user's viewpoint. In this paper, we conduct a subjective study to investigate users' quality perception of 3D meshes with dynamic Level of Detail in a Virtual Reality environment. The subjective experiment is carried out with five 3D meshes of different characteristics, four Levels of Detail, and four distance settings. The results of the experiment show that the impact of the dynamic level of detail depends on both the position of the 3D object in the virtual world and the number of vertices of the original mesh. In addition, we present a quality model that can accurately predict the MOS score of a LoD version of a 3D mesh from the number of vertices and the distance from the viewpoint.