Universal Beta Splatting

2025-09-30

AI Summary

This work addresses the challenge of jointly modeling spatial, view-dependent, and temporal dependencies in explicit radiance field representations. We propose Universal Beta Splatting (UBS), which generalizes 3D Gaussian splatting to N-dimensional anisotropic Beta kernels—establishing the first explicit radiance field framework supporting controllable cross-dimensional dependency modeling. Our key contribution is the introduction of learnable Beta kernels, enabling joint anisotropic appearance and dynamic scene modeling without auxiliary networks or additional color encoding; all parameters are physically interpretable and decomposable into surface, texture, diffuse, specular, and static/dynamic components. Implemented via a CUDA-accelerated N-dimensional renderer compatible with Gaussian splatting, UBS supports real-time rendering and end-to-end optimization. Extensive experiments demonstrate state-of-the-art performance on static, view-dependent, and dynamic scenes, validating Beta kernels as a universal, scalable primitive for explicit radiance field modeling.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Kernel Methods

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

We introduce Universal Beta Splatting (UBS), a unified framework that generalizes 3D Gaussian Splatting to N-dimensional anisotropic Beta kernels for explicit radiance field rendering. Unlike fixed Gaussian primitives, Beta kernels enable controllable dependency modeling across spatial, angular, and temporal dimensions within a single representation. Our unified approach captures complex light transport effects, handles anisotropic view-dependent appearance, and models scene dynamics without requiring auxiliary networks or specific color encodings. UBS maintains backward compatibility by approximating to Gaussian Splatting as a special case, guaranteeing plug-in usability and lower performance bounds. The learned Beta parameters naturally decompose scene properties into interpretable without explicit supervision: spatial (surface vs. texture), angular (diffuse vs. specular), and temporal (static vs. dynamic). Our CUDA-accelerated implementation achieves real-time rendering while consistently outperforming existing methods across static, view-dependent, and dynamic benchmarks, establishing Beta kernels as a scalable universal primitive for radiance field rendering. Our project website is available at https://rongliu-leo.github.io/universal-beta-splatting/.

Problem

• Generalizing Gaussian Splatting to N-dimensional Beta kernels
• Modeling dependencies across spatial, angular, and temporal dimensions
• Capturing complex light transport and dynamic scene effects

Innovation

• Generalizes Gaussian Splatting with N-dimensional Beta kernels
• Models spatial, angular, and temporal dependencies controllably
• Achieves real-time rendering without auxiliary networks or encodings