RAP: Fast Feedforward Rendering-Free Attribute-Guided Primitive Importance Score Prediction for Efficient 3D Gaussian Splatting Processing

TL;DR

RAP addresses the inefficiency of 3D Gaussian Splatting by predicting per-primitive importance without rendering. It uses a 15-dimensional intrinsic-attribute feature vector and a lightweight MLP to produce scores in $S_i \in [0,1]$, trained with three losses to preserve fidelity while enabling pruning. The method is fast, rendering-free at inference, and generalizes across unseen scenes, improving pruning, compression, and transmission performance. Extensive experiments on Mip-NeRF360, DeepBlending, and Tanks & Temples show RAP outperforms rendering-based and gradient-based baselines, enabling substantial storage savings with minimal quality loss.

Abstract

3D Gaussian Splatting (3DGS) has emerged as a leading technology for high-quality 3D scene reconstruction. However, the iterative refinement and densification process leads to the generation of a large number of primitives, each contributing to the reconstruction to a substantially different extent. Estimating primitive importance is thus crucial, both for removing redundancy during reconstruction and for enabling efficient compression and transmission. Existing methods typically rely on rendering-based analyses, where each primitive is evaluated through its contribution across multiple camera viewpoints. However, such methods are sensitive to the number and selection of views, rely on specialized differentiable rasterizers, and have long calculation times that grow linearly with view count, making them difficult to integrate as plug-and-play modules and limiting scalability and generalization. To address these issues, we propose RAP, a fast feedforward rendering-free attribute-guided method for efficient importance score prediction in 3DGS. RAP infers primitive significance directly from intrinsic Gaussian attributes and local neighborhood statistics, avoiding rendering-based or visibility-dependent computations. A compact MLP predicts per-primitive importance scores using rendering loss, pruning-aware loss, and significance distribution regularization. After training on a small set of scenes, RAP generalizes effectively to unseen data and can be seamlessly integrated into reconstruction, compression, and transmission pipelines. Our code is publicly available at https://github.com/yyyykf/RAP.

Figures (9)

• Figure 1: Visualization of redundant and low-importance Gaussians in the Train scene from Tanks&Temples.
• Figure 2: RAP learning framework.
• Figure 3: PSNR vs. retention ratio curves.
• Figure 4: R-D curves for GSC with 20% pruning
• Figure 5: Ablation studies of the proposed feature and loss.