Oct 2026· Proceedings of the 31st ACM International Conference on 3D Web Technology· pp. 1-5· 0 citations· 11 references
TL;DR
Experimental results show that the proposed framework improves scene scalability, increases the number of rendered mesh by 8 × while maintaining comparable runtime performance to conventional approaches and achieving better perceptual quality under varying GPU memory constraints.
Abstract
Dynamic 3D meshes are crucial for immersive Virtual Reality (VR) applications, but rendering multiple high-quality meshes on WebVR environments is challenging due to limited GPU memory and computational resources. This paper presents a QoE-aware adaptive rendering framework that jointly considers perceptual quality, geometric complexity, and GPU memory constraints for Level of Detail (LoD) allocation. The LoD selection is formulated as a constrained optimization problem to maximize overall QoE under available resource budgets. Experimental results on a Meta Quest 2 headset show that the proposed framework improves scene scalability, increases the number of rendered mesh by 8 × while maintaining comparable runtime performance to conventional approaches and achieving better perceptual quality under varying GPU memory constraints.
Geometry-Aware Redundancy Optimization (GARO) is proposed, a unified redundancy measurement framework in the adaptive density control stage of the traditional dynamic scene reconstruction pipeline that achieves robust trade-offs between quality and speed.
Hui-Wen Xue, Kai-Xing Zhao, Zuheng Ming et al.· IEEE International Conferenc...· 0 citations
This article proposes an efits of the proposed approach in terms of both execution time and reconstruction quality even in very low throughput scenarios of bit-per-vertex-per-frame (bpvf).
Gerasimos Arvanitis, A. Lalos, Konstantinos Moustakas· 0 citations
This study proposes an optimized AI-driven 3D reconstruction pipeline featuring: (1) improved cGAN with multi-scale perceptual and edge-aware losses; (2) adaptive LOD algorithm for real-time WebGL rendering; (3) hybrid neural rendering with progressive streaming. Experiments on 12 cultural relics demonstrate 28.5% impr...
Traditional 3D representation models, such as 3D meshes and 3D point clouds, provide accurate geometric measurements but lack the visual fidelity of modern novel view synthesis approaches. In contrast, recent differentiable rendering methods, such as Plenoxels and Gaussian Splatting, enable photorealistic real-time vis...
Raphael Kunert, R. Richter, Jürgen Döllner· Proceedings of the 31st ACM...· 0 citations
Introduction: Pixel art has been explored since the early development of 2D computer graphics and remains a popular aesthetic, particularly in video games. However, reproducing this style in real-time 3D environments remains challenging. Objective: To develop a shader that emulates pixel art aesthetics in real-time 3D...
Naomi Celestino Ribes, Anderson Maciel, R. Torchelsen· Brazilian Symposium on Games...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.