MipSLAM: Alias-Free Gaussian Splatting SLAM

Yingzhao Li; Yan Li; Shixiong Tian; Yanjie Liu; Lijun Zhao; Gim Hee Lee

cs.CVMar 7, 2026

MipSLAM: Alias-Free Gaussian Splatting SLAM

Authors: Yingzhao Li, Yan Li, Shixiong Tian, Yanjie Liu, Lijun Zhao, Gim Hee Lee

Organizations: State Key Laboratory of Robotics and Systems (HIT), Harbin Institute of Technology. · Yangtze River Delta HlT Robot Technology Research Institute. · Department of Computer Science, National University of Singapore.

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. Extensive evaluations on Replica and TUM datasets demonstrate that MipSLAM achieves state-of-the-art rendering quality and localization accuracy across multiple resolutions. Code is available at https://github.com/yzli1998/MipSLAM.

Figures & tables

Panel of Figure 1 — Fig. 1 : Our method reformulates pixel rendering as a continuous integration process via stratified importance sampling, ensuring antialiasing quality and computational efficiency. The dashed ellipses depict Gaussian principal axes, color-coded by condition number: red (high), orange (moderate), purple (boundary enhancement), and green (low complexity).

Fig. 2 : MipSLAM tracks RGB-D streams, estimates per-frame poses, selects keyframes, and projects Gaussian primitives into a global map. It incrementally computes 3D filter scales, compares rendered and GT RGB-D in a sliding window, and jointly optimizes poses and maps via gradient backpropagation. For $\alpha$ -blending, we use elliptical projection and condition number analysis to reduce aliasing via numerical integration. SA-PGO corrects localization drift from anti-aliasing.

Panel of Figure 3 — Fig. 3 : SA-PGO leverages frequency-domain trajectory analysis and graph spectral theory to achieve adaptive noise suppression and intelligent spectral-aware optimization.

Method		PSNR $\uparrow$						SSIM $\uparrow$						LPIPS $\downarrow$
		2	1	$1/2$	$1/4$	$1/8$	Avg	2	1	$1/2$	$1/4$	$1/8$	Avg	2	1	$1/2$	$1/4$	$1/8$	Avg
Recon	3D GS [ 11 ]	38.71	42.20	38.22	29.84	25.99	34.99	0.974	0.981	0.963	0.888	0.835	0.928	0.108	0.083	0.069	0.091	0.121	0.094
	Sca-GS [ 15 ]	38.42	40.97	35.86	29.91	24.28	33.89	0.975	0.968	0.954	0.884	0.796	0.915	0.101	0.085	0.067	0.079	0.137	0.094
SLAM	SplaTAM [ 10 ]	33.68	35.54	32.80	25.22	22.64	29.98	0.974	0.976	0.966	0.912	0.842	0.934	0.124	0.073	0.064	0.127	0.123	0.102
	MonoGS [ 16 ]	36.98	39.51	34.67	28.43	24.01	32.72	0.969	0.975	0.963	0.899	0.812	0.924	0.077	0.044	0.031	0.071	0.098	0.064
	GS-ICP [ 8 ]	34.00	37.55	33.64	28.68	24.67	31.71	0.952	0.967	0.950	0.870	0.787	0.905	0.074	0.056	0.044	0.082	0.104	0.072

TABLE I: Multiresolution testing results on the Replica dataset [ 23 ] . The best results are shown in red, the second-best results in blue, and the third-best results in green. The data are averages over 8 sequences. A dash (‘–’) indicates a system crash during tracking.

Method	PSNR $\uparrow$						SSIM $\uparrow$						LPIPS $\downarrow$
	4	2	1	$1/2$	$1/4$	Avg	4	2	1	$1/2$	$1/4$	Avg	4	2	1	$1/2$	$1/4$	Avg
SplaTAM [ 10 ]	21.70	21.78	21.93	21.46	20.68	21.51	0.821	0.822	0.785	0.765	0.682	0.775	0.302	0.289	0.244	0.222	0.258	0.263
MonoGS [ 16 ]	23.04	23.33	23.57	21.44	20.45	22.37	0.862	0.823	0.784	0.783	0.742	0.799	0.275	0.273	0.247	0.193	0.147	0.227
GS-ICP [ 8 ]	17.67	17.68	17.69	17.57	17.27	17.58	0.851	0.780	0.706	0.659	0.646	0.728	0.307	0.317	0.302	0.282	0.234	0.288
Ours	23.34	23.52	23.21	21.76	22.03	22.77	0.865	0.827	0.782	0.789	0.788	0.810	0.271	0.268	0.249	0.189	0.123	0.220

TABLE II: Multiresolution testing results on the TUM dataset [ 24 ] . The best results are shown in red, the second-best results in blue, and the third-best results in green.

Method		R0	R1	R2	Off0	Off1	Off2	Off3	Off4	Avg
Implicit	NICE-SLAM [ 29 ]	0.97	1.31	1.07	0.88	1.00	1.06	1.10	1.13	1.07
	Point-SLAM [ 21 ]	0.61	0.41	0.37	0.38	0.48	0.54	0.69	0.72	0.53
	GO-SLAM [ 28 ]	0.34	0.29	0.30	0.32	0.30	0.39	0.39	0.35	0.35
3D GS	MonoGS [ 16 ]	0.48	0.32	0.33	0.51	0.57	0.22	0.18	2.06	0.58
	SplaTAM [ 10 ]	0.31	0.40	0.29	0.47	0.27	0.29	0.32	0.72	0.38
	Photo-SLAM [ 9 ]	0.54	0.39	0.31	0.52	0.44	1.28	0.78	0.58	0.61

TABLE III : Comparison of localization accuracy (ATE RMSE [cm]

\downarrow

) on the Replica dataset [ 23 ] .

Fig. 4 : Testing results at 1 $/$ 4 resolution on the Replica [ 23 ] and TUM datasets [ 24 ] . Previous methods have exhibited varying degrees of aliasing, blurring, and inflation issues. Our MipSLAM can faithfully represent the real world.

Fig. 5 : Testing results at 2 $\times$ resolution on the Replica [ 23 ] and TUM datasets [ 24 ] . Other methods exhibit a decline in local regions, whereas our MipSLAM is capable of representing the detailed textures.

3DGS SLAM

3D Gaussian Splatting simultaneous localization and mapping builds renderable 3D Gaussian maps while estimating camera motion online.

See all 63 papers

MMD-SLAM: Structure-Enhanced Multi-Meta Gaussian Distribution-Guided Visual SLAM
Jun 18, 2026Fan Zhu, Ziyu Chen, Peichen Liu +6Simultaneous Localization and Mapping 3D Gaussian Splatting
HFIPS, Chinese Academy of Sciences, Hefei, China. · University of Science and Technology of China, Hefei, China. · Aarhus University, Aarhus, Denmark. +3
RMGS-SLAM: Real-time Multi-sensor Gaussian Splatting SLAM
Apr 14, 2026Dongen Li, Yi Liu, Junqi Liu +8Simultaneous Localization and Mapping 3DGS SLAM
Dongen Li, Yi Liu, Zewen Sun, Zefan Huang, Chengran Yuan, Francis E.H. Tay, and Marcelo H. Ang Jr. are with the Advanced Robotics Centre, National University of Singapore. · Junqi Liu and Hongliang Guo are with the College of Computer Science, Sichuan University. · Shuo Sun is with Singapore-MIT Alliance for Research and Technology. +1
Geometry-Aware Online Mapping for 3D Gaussian Splatting SLAM
Aug 14, 2026Thai Luu, Quan Tran, Hieu Phan +13DGS SLAM 3D Gaussian Splatting
Cognitive Robotics Lab, Department of Electrical Engineering and Computer Science, University of Arkansas, USA. · Center for AI Research, VinUniversity, Ha Noi, Vietnam.

Jun 18, 2026J/K move · Enter open · S save

MMD-SLAM: Structure-Enhanced Multi-Meta Gaussian Distribution-Guided Visual SLAM

Fan Zhu, Ziyu Chen, Peichen Liu, Yifan Zhao+5

HFIPS, Chinese Academy of Sciences, Hefei, China. · University of Science and Technology of China, Hefei, China. · Aarhus University, Aarhus, Denmark.+3

Simultaneous Localization and Mapping 3D Gaussian Splatting 3DGS SLAM Visual SLAM

Open paper AskPDF arXiv

3D Gaussian Splatting (3DGS) has significantly boosted novel view synthesis and high-fidelity scene reconstruction, expanding the potential of 3DGS-based Visual Simultaneous Localization and Mapping (SLAM) methods. However, most existing systems fail to fully exploit the underlying structural information, which limits rendering quality and often leads to inconsistent maps. To address these limitations, we propose MMD-SLAM, a structure-enhanced Visual SLAM framework that leverages the Atlanta World (AW) assumption to guide a Multi-Meta Gaussian representation for photorealistic mapping. First, we introduce a point-line fusion strategy for pose optimization, where 3D line segments are incorporated to improve tracking robustness and provide additional constraints for mapping. Second, we design a Multi-Meta Gaussian representation with dominant directions, explicitly encoding structural priors from the AW hypothesis. Finally, we propose a Gaussian evolution strategy that adapts to scene geometry and incorporates structural cues into global optimization. Extensive experiments demonstrate that these innovations enable MMD-SLAM to achieve state-of-the-art performance in both tracking accuracy and mapping quality. e.g., our method achieves a 48.56% reduction in ATE RMSE on ScanNet and a 5.71% improvement in PSNR on Replica, compared with MonoGS.

Abstract

3DGS SLAM

Explore similar work

MMD-SLAM: Structure-Enhanced Multi-Meta Gaussian Distribution-Guided Visual SLAM

RMGS-SLAM: Real-time Multi-sensor Gaussian Splatting SLAM

Geometry-Aware Online Mapping for 3D Gaussian Splatting SLAM