MipSLAM: Alias-Free Gaussian Splatting SLAM

2026-03-07

AI Summary

This work proposes the first frequency-aware 3D Gaussian splatting SLAM framework to address the susceptibility of existing systems to aliasing artifacts and trajectory drift under varying camera configurations, which often compromises the trade-off between novel view synthesis quality and pose estimation robustness. The approach introduces an Elliptical Adaptive Anti-aliasing (EAA) algorithm for geometry-aware anti-aliased rendering and a Spectral-Aware Pose Graph Optimization (SA-PGO) module that suppresses high-frequency noise and drift in the frequency domain. Coupled with a local frequency-aware loss function, the method significantly enhances geometric detail recovery and localization stability. Experiments demonstrate that the system achieves state-of-the-art performance in both novel view synthesis and pose accuracy on the Replica and TUM datasets, while supporting real-time operation across multiple resolutions.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper introduces MipSLAM, a frequency-aware 3D Gaussian Splatting (3DGS) SLAM framework capable of high-fidelity anti-aliased novel view synthesis and robust pose estimation under varying camera configurations. Existing 3DGS-based SLAM systems often suffer from aliasing artifacts and trajectory drift due to inadequate filtering and purely spatial optimization. To overcome these limitations, we propose an Elliptical Adaptive Anti-aliasing (EAA) algorithm that approximates Gaussian contributions via geometry-aware numerical integration, avoiding costly analytic computation. Furthermore, we present a Spectral-Aware Pose Graph Optimization (SA-PGO) module that reformulates trajectory estimation in the frequency domain, effectively suppressing high-frequency noise and drift through graph Laplacian analysis. A novel local frequency-domain perceptual loss is also introduced to enhance fine-grained geometric detail recovery. Extensive evaluations on Replica and TUM datasets demonstrate that MipSLAM achieves state-of-the-art rendering quality and localization accuracy across multiple resolutions while maintaining real-time capability. Code is available at https://github.com/yzli1998/MipSLAM.

Problem

• aliasing artifacts
• trajectory drift
• 3D Gaussian Splatting
• SLAM
• novel view synthesis

Innovation

• Gaussian Splatting
• Anti-aliasing
• Frequency-aware SLAM
• Pose Graph Optimization
• Novel View Synthesis