RayGaussX: Accelerating Gaussian-Based Ray Marching for Real-Time and High-Quality Novel View Synthesis

RayGauss has achieved state-of-the-art rendering quality for novel-view synthesis on synthetic and indoor scenes by representing radiance and density fields with irregularly distributed elliptical basis functions, rendered via volume ray casting using a Bounding Volume Hierarchy (BVH). However, its computational cost prevents real-time rendering on real-world scenes. Our approach, RayGaussX, builds on RayGauss by introducing key contributions that accelerate both training and inference. Specifically, we incorporate volumetric rendering acceleration strategies such as empty-space skipping and adaptive sampling, enhance ray coherence, and introduce scale regularization to reduce false-positive intersections. Additionally, we propose a new densification criterion that improves density distribution in distant regions, leading to enhanced graphical quality on larger scenes. As a result, RayGaussX achieves $5 \times$ to $12 \times$ faster training and $50 \times$ to $80 \times$ higher rendering speeds (FPS) on real-world datasets while improving visual quality by up to $+0.56 d B$ in PSNR. The associated code is available at: github.com/hugobl1/raygaussx.