Z-Order Transformer for Feed-Forward Gaussian Splatting

TL;DR

This paper introduces a transformer-based approach with a Z-order strategy for efficient, high-quality, feed-forward Gaussian Splatting in 3D view synthesis, reducing primitives and improving speed.

Contribution

It presents a novel transformer architecture with a Z-order organization to capture spatial relationships, suppress redundancy, and predict Gaussian attributes in a single pass.

Findings

Achieves faster, high-quality view synthesis with fewer Gaussian primitives.

Effectively captures spatial and semantic relationships among Gaussians.

Reduces redundancy while preserving structural details.

Abstract

Recent advances in 3D Gaussian Splatting (3DGS) have enabled significant progress in photorealistic novel view synthesis. However, traditional 3DGS relies on a slow, iterative optimization process, which limits its use in scenarios demanding real-time results. To overcome this bottleneck, recent feed-forward methods aim to predict Gaussian attributes directly from images, but they often struggle with the redundancy of Gaussian primitives and rendering quality. In this work, we introduce a transformer-based architecture specifically designed for feed-forward Gaussian Splatting. Our key insight is that spatial and semantic relationships among Gaussians can be effectively captured through a sparse attention mechanism, enabled by a Z-order strategy that organizes the unstructured Gaussian set into a spatially coherent sequence. Furthermore, we incorporate this Z-order strategy to adaptively suppress redundancy while preserving critical structural details. This allows the transformer to efficiently model context, compress Gaussian primitives, and predict Gaussian attributes in a single forward pass. Comprehensive experiments demonstrate that our method achieves fast and high-quality novel view synthesis with fewer Gaussian primitives.