AMB3R-SLAM: Kilometer-scale SLAM with Hierarchical Backend
Not on the map. Why?
Abstract (arXiv)
We present AMB3R-SLAM, a real-time monocular SLAM system capable of reconstructing kilometer-scale trajectories over 10k frames on a single consumer-grade GPU. Our model couples a lightweight front-end for low-latency online tracking with a hierarchical backend that progressively enforces local, mid-level, and global consistency. By avoiding bundle adjustment that relies on the static world assumption, our system naturally handles complex dynamic scenes out of the box. Furthermore, we demonstrate that our method can be extended to leverage stereo, RGB-D, and LiDAR as additional inputs. AMB3R-SLAM achieves strong camera tracking performance across 9 datasets, reducing the absolute trajectory error (ATE) of previous state-of-the-art methods on VBR and Oxford Spires by over 70%. With additional LiDAR input, our model further reduces ATE to sub-meter level on KITTI and VBR datasets.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 72 of its 90 references that are in the map. First posted on arXiv 17 Sep 2026.
Builds on
72 of 90 references.
References · 72
- 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,567
- 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
- 2015ORB-SLAM: a Versatile and Accurate Monocular SLAM Systemcited by 988
- 2024Grounding Image Matching in 3D with MASt3Rcited by 983
- 2016ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Camerascited by 806
- 2025Depth Anything 3: Recovering the Visual Space from Any Viewscited by 692
- 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 687
- 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 589
- 2025Continuous 3D Perception Model with Persistent Statecited by 548
- 2016Direct Sparse Odometrycited by 485
- 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
- 2024MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motioncited by 442
38 more
- 2023Gaussian Splatting SLAMcited by 415
- 2025Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Passcited by 367
- 2025MapAnything: Universal Feed-Forward Metric 3D Reconstructioncited by 314
- 20243D Reconstruction with Spatial Memorycited by 277
- 2024MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videoscited by 228
- 2024Global Structure-from-Motion Revisitedcited by 218
- 2024MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priorscited by 205
- 2022Deep Patch Visual Odometrycited by 190
- 2020LIO-SAM: Tightly-coupled Lidar Inertial Odometry via Smoothing and Mappingcited by 188
- 2023Co-SLAM: Joint Coordinate and Sparse Parametric Encodings for Neural Real-Time SLAMcited by 184
- 2019ReFusion: 3D Reconstruction in Dynamic Environments for RGB-D Cameras Exploiting Residualscited by 181
- 2025ViPE: Video Pose Engine for 3D Geometric Perceptioncited by 170
- 2025Streaming 4D Visual Geometry Transformercited by 161
- 2023Gaussian-SLAM: Photo-realistic Dense SLAM with Gaussian Splattingcited by 154
- 2023Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular, Stereo, and RGB-D Camerascited by 141
- 2024MV-DUSt3R+: Single-Stage Scene Reconstruction from Sparse Views In 2 Secondscited by 141
- 2025MUSt3R: Multi-view Network for Stereo 3D Reconstructioncited by 141
- 2025VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifoldcited by 136
- 2022KISS-ICP: In Defense of Point-to-Point ICP -- Simple, Accurate, and Robust Registration If Done the Right Waycited by 132
- 2025TTT3R: 3D Reconstruction as Test-Time Trainingcited by 129
- 2025VGGT-Long: Chunk it, Loop it, Align it -- Pushing VGGT's Limits on Kilometer-scale Long RGB Sequencescited by 115
- 2024SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videoscited by 102
- 2022LaMAR: Benchmarking Localization and Mapping for Augmented Realitycited by 100
- 2025Aether: Geometric-Aware Unified World Modelingcited by 97
- 2025Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memorycited by 87
- 2025STream3R: Scalable Sequential 3D Reconstruction with Causal Transformercited by 82
- 2025Pow3R: Empowering Unconstrained 3D Reconstruction with Camera and Scene Priorscited by 72
- 2024Deep Patch Visual SLAMcited by 71
- 2023vMAP: Vectorised Object Mapping for Neural Field SLAMcited by 67
- 2025Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstructioncited by 67
- 2025WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environmentscited by 50
- 2025Light3R-SfM: Towards Feed-forward Structure-from-Motioncited by 47
- 2025Dens3R: A Foundation Model for 3D Geometry Predictioncited by 46
- 2021LVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry via Smoothing and Mappingcited by 45
- 2025LONG3R: Long Sequence Streaming 3D Reconstructioncited by 45
- 2024PIN-SLAM: LiDAR SLAM Using a Point-Based Implicit Neural Representation for Achieving Global Map Consistencycited by 42
- 2026LoGeR: Long-Context Geometric Reconstruction with Hybrid Memorycited by 39
- 2025WinT3R: Window-Based Streaming Reconstruction with Camera Token Poolcited by 35
50 most cited of 72 All 72 (JSON)
Built on by
Not cited yet.