USP-Gaussian: Unifying Spike-based Image Reconstruction, Pose Correction and Gaussian Splatting

TL;DR

USP-Gaussian introduces an end-to-end framework that unifies spike-based image reconstruction, pose correction, and Gaussian splatting, significantly improving 3D reconstruction accuracy and robustness over previous cascaded approaches.

Contribution

It presents a novel unified optimization framework that integrates spike-to-image reconstruction, pose correction, and 3D Gaussian splatting into a single end-to-end system.

Findings

Outperforms previous methods on synthetic datasets with accurate poses.

Effectively reduces noise and preserves fine textures in real-world scenarios.

Eliminates cascading errors by joint optimization.

Abstract

Spike cameras, as an innovative neuromorphic camera that captures scenes with the 0-1 bit stream at 40 kHz, are increasingly employed for the 3D reconstruction task via Neural Radiance Fields (NeRF) or 3D Gaussian Splatting (3DGS). Previous spike-based 3D reconstruction approaches often employ a casecased pipeline: starting with high-quality image reconstruction from spike streams based on established spike-to-image reconstruction algorithms, then progressing to camera pose estimation and 3D reconstruction. However, this cascaded approach suffers from substantial cumulative errors, where quality limitations of initial image reconstructions negatively impact pose estimation, ultimately degrading the fidelity of the 3D reconstruction. To address these issues, we propose a synergistic optimization framework, \textbf{USP-Gaussian}, that unifies spike-based image reconstruction, pose correction, and Gaussian splatting into an end-to-end framework. Leveraging the multi-view consistency afforded by 3DGS and the motion capture capability of the spike camera, our framework enables a joint iterative optimization that seamlessly integrates information between the spike-to-image network and 3DGS. Experiments on synthetic datasets with accurate poses demonstrate that our method surpasses previous approaches by effectively eliminating cascading errors. Moreover, we integrate pose optimization to achieve robust 3D reconstruction in real-world scenarios with inaccurate initial poses, outperforming alternative methods by effectively reducing noise and preserving fine texture details. Our code, data and trained models will be available at https://github.com/chenkang455/USP-Gaussian.