3DGEER: Exact and Efficient Volumetric Rendering with 3D Gaussians

2025-05-29

AI Summary

3D Gaussian Splatting (3DGS) suffers from severe rendering quality degradation under wide-field-of-view (FoV) cameras due to its reliance on approximating 3D Gaussians as 2D Gaussians via perspective projection. This work introduces the first physically exact, real-time voxelized Gaussian rendering framework. We derive a closed-form analytical solution for the ray-integrated density of 3D Gaussians based on physical principles; propose Particle-Bounded Frustum (PBF) for efficient visibility culling; and design Bi-polar Equiangular Projection (BEAP) to unify pinhole and fisheye camera models under a single geometric formulation. Crucially, our method eliminates all 2D projection approximations while enabling differentiable voxel rendering at real-time rates. Extensive evaluation across diverse FoV datasets demonstrates consistent superiority over existing 3DGS variants, achieving state-of-the-art performance in both reconstruction fidelity and computational efficiency.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures

Abstract

3D Gaussian Splatting (3DGS) marks a significant milestone in balancing the quality and efficiency of differentiable rendering. However, its high efficiency stems from an approximation of projecting 3D Gaussians onto the image plane as 2D Gaussians, which inherently limits rendering quality--particularly under large Field-of-View (FoV) camera inputs. While several recent works have extended 3DGS to mitigate these approximation errors, none have successfully achieved both exactness and high efficiency simultaneously. In this work, we introduce 3DGEER, an Exact and Efficient Volumetric Gaussian Rendering method. Starting from first principles, we derive a closed-form expression for the density integral along a ray traversing a 3D Gaussian distribution. This formulation enables precise forward rendering with arbitrary camera models and supports gradient-based optimization of 3D Gaussian parameters. To ensure both exactness and real-time performance, we propose an efficient method for computing a tight Particle Bounding Frustum (PBF) for each 3D Gaussian, enabling accurate and efficient ray-Gaussian association. We also introduce a novel Bipolar Equiangular Projection (BEAP) representation to accelerate ray association under generic camera models. BEAP further provides a more uniform ray sampling strategy to apply supervision, which empirically improves reconstruction quality. Experiments on multiple pinhole and fisheye datasets show that our method consistently outperforms prior methods, establishing a new state-of-the-art in real-time neural rendering.

Problem

• Achieves exact volumetric rendering with 3D Gaussians efficiently
• Overcomes approximation errors in 3DGS under large FoV inputs
• Supports arbitrary camera models and real-time performance

Innovation

• Closed-form density integral for exact rendering
• Particle Bounding Frustum for efficient ray-Gaussian association
• Bipolar Equiangular Projection for uniform ray sampling