RaRa Clipper: A Clipper for Gaussian Splatting Based on Ray Tracer and Rasterizer

2025-06-25

AI Summary

In Gaussian splatting, the volumetric nature of primitives renders conventional hard clipping inadequate for pixel-accurate culling. To address this, we propose RaRa—a hybrid rasterization-ray-tracing rendering framework. First, rasterization rapidly identifies and culls Gaussians intersecting the clipping plane; then, ray tracing computes continuous attenuation weights under partial occlusion, enabling sub-pixel contribution modeling and smooth, anti-aliased clipping. RaRa supports complex Gaussian scenes—including multi-layered structures and hair—balancing computational efficiency with geometric and photometric fidelity. Extensive evaluation across multiple datasets demonstrates that RaRa preserves high fidelity in unclipped regions while achieving real-time performance (>30 FPS). Quantitatively, it surpasses state-of-the-art methods in PSNR and SSIM; qualitatively, it delivers superior visual quality with seamless clipping boundaries and accurate occlusion handling.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Hardware-aware ML

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Security and Privacy: Data transparency and provenance

Abstract

With the advancement of Gaussian Splatting techniques, a growing number of datasets based on this representation have been developed. However, performing accurate and efficient clipping for Gaussian Splatting remains a challenging and unresolved problem, primarily due to the volumetric nature of Gaussian primitives, which makes hard clipping incapable of precisely localizing their pixel-level contributions. In this paper, we propose a hybrid rendering framework that combines rasterization and ray tracing to achieve efficient and high-fidelity clipping of Gaussian Splatting data. At the core of our method is the RaRa strategy, which first leverages rasterization to quickly identify Gaussians intersected by the clipping plane, followed by ray tracing to compute attenuation weights based on their partial occlusion. These weights are then used to accurately estimate each Gaussian's contribution to the final image, enabling smooth and continuous clipping effects. We validate our approach on diverse datasets, including general Gaussians, hair strand Gaussians, and multi-layer Gaussians, and conduct user studies to evaluate both perceptual quality and quantitative performance. Experimental results demonstrate that our method delivers visually superior results while maintaining real-time rendering performance and preserving high fidelity in the unclipped regions.

Problem

• Achieving accurate clipping for Gaussian Splatting data
• Combining rasterization and ray tracing for efficient clipping
• Preserving high fidelity in unclipped regions during rendering

Innovation

• Hybrid rendering with rasterization and ray tracing
• RaRa strategy for efficient Gaussian identification
• Attenuation weights for precise contribution estimation