AI & ML Efficiency Breakthrough

Mobile-GS achieves real-time Gaussian Splatting on mobile devices by replacing the sorting-based alpha-blending bottleneck with depth-aware order-independent rendering.

arXiv · March 13, 2026 · 2603.11531

Xiaobiao Du, Yida Wang, Kun Zhan, Xin Yu

Why it matters

Rendering 3DGS on mobile is typically stalled by depth sorting. This method uses a neural view-dependent enhancement to fix transparency artifacts from unsorted rendering, making high-fidelity 3D radiance fields practical for mobile XR applications.

From the abstract

3D Gaussian Splatting (3DGS) has emerged as a powerful representation for high-quality rendering across a wide range ofthis http URL, its high computational demands and large storage costs pose significant challenges for deployment on mobile devices. In this work, we propose a mobile-tailored real-time Gaussian Splatting method, dubbed Mobile-GS, enabling efficient inference of Gaussian Splatting on edge devices. Specifically, we first identify alpha blending as the primary computational bottlen