Immediate 3D Gaussian Splat Reconstruction of Unordered Input with Global Consistency

2026-07-15

AI Summary

This work addresses the challenges of real-time 3D reconstruction from unordered image sequences, where existing methods struggle to balance immediate feedback with global consistency. We propose the first instant 3D Gaussian Splatting (3DGS) framework capable of handling unordered inputs, leveraging a co-visibility graph–driven fast frame matching strategy, cluster-based loop detection, and a non-sequential loop closure mechanism within a progressive hierarchical reconstruction architecture. By integrating visual place recognition, GPU-accelerated optimization, and intelligent Gaussian primitive placement, our approach enables efficient, globally consistent radiance field construction. Extensive experiments on datasets comprising thousands of images demonstrate high-quality, low-latency reconstruction, significantly improving scalability and accuracy for large-scale scenes.

Abstract

3D Gaussian Splatting (3DGS) has become the method of choice for reconstructing and real-time rendering of captured scenes. To capture a scene with good visual quality, continuous image sequences are usually combined with out-of-order shots for better scene coverage. Structure from motion can reconstruct such captures, but only after they are all available and often with high computational cost. Incremental reconstruction methods -- often derived from SLAM solutions -- provide immediate feedback, but cannot handle the out-of-order capture we require. We provide the first immediate feedback solution for such radiance field capture that provides global consistency. We first introduce a method for fast matching in out-of-order sequences, by repurposing visual place recognition models and a covisibility graph, and provide an efficient way to find highly connected keyframes, improving quality even for ordered sequences. We show how these steps -- together with GPU optimization and careful Gaussian primitive placement -- provide fast local reconstruction, in our challenging radiance field reconstruction case. We then introduce a novel cluster-based method, again using the covisibility graph, to provide efficient loop closure that does not require sequential input. Finally, to handle large scenes in our context, we introduce a progressive hierarchy that allows our method to scale to large environments, without compromising efficiency. Our results show we provide immediate feedback 3DGS reconstruction with good visual quality in several datasets, with up to thousands of input images.

Problem

• 3D Gaussian Splatting
• unordered input
• global consistency
• immediate reconstruction
• radiance field

Innovation

• 3D Gaussian Splatting
• out-of-order reconstruction
• global consistency
• covisibility graph
• incremental radiance field