# A Minimal Subset Approach for Informed Keyframe Sampling in Large-Scale SLAM

arXiv: https://arxiv.org/abs/2501.01791 · HTML: https://mlsanity.com/paper/2501.01791/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 33 references that are in the map.
- First posted on arXiv 3 Jan 2025.

## Builds on (most cited first, 15 of 15)

- [Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Age](https://mlsanity.com/paper/1606.05830/) (2016)
- [LIO-SAM: Tightly-coupled Lidar Inertial Odometry via Smoothing and Mapping](https://mlsanity.com/paper/2007.00258/) (2020)
- [KISS-ICP: In Defense of Point-to-Point ICP -- Simple, Accurate, and Robust Registration If Done the Right Way](https://mlsanity.com/paper/2209.15397/) (2022)
- [Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environments](https://mlsanity.com/paper/2109.13494/) (2021)
- [OverlapTransformer: An Efficient and Rotation-Invariant Transformer Network for LiDAR-Based Place Recognition](https://mlsanity.com/paper/2203.03397/) (2022)
- [LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM](https://mlsanity.com/paper/2103.05056/) (2021)
- [The Newer College Dataset: Handheld LiDAR, Inertial and Vision with Ground Truth](https://mlsanity.com/paper/2003.05691/) (2020)
- [A Survey on Global LiDAR Localization: Challenges, Advances and Open Problems](https://mlsanity.com/paper/2302.07433/) (2023)
- [Direct LiDAR Odometry: Fast Localization with Dense Point Clouds](https://mlsanity.com/paper/2110.00605/) (2021)
- [RING++: Roto-translation Invariant Gram for Global Localization on a Sparse Scan Map](https://mlsanity.com/paper/2210.05984/) (2022)
- [LAMP 2.0: A Robust Multi-Robot SLAM System for Operation in Challenging Large-Scale Underground Environments](https://mlsanity.com/paper/2205.13135/) (2022)
- [Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM](https://mlsanity.com/paper/2205.12402/) (2022)
- [FRAME: A Modular Framework for Autonomous Map Merging: Advancements in the Field](https://mlsanity.com/paper/2404.18006/) (2024)
- [Sparse Pose Graph Optimization in Cycle Space](https://mlsanity.com/paper/2203.15597/) (2022)
- [HiPE: Hierarchical Initialization for Pose Graphs](https://mlsanity.com/paper/2207.01582/) (2022)

## Built on by (most cited first, 2 of 2)

- [Automatic Map Density Selection for Locally-Performant Visual Place Recognition](https://mlsanity.com/paper/2602.21473/) (2026)
- [InLiER: Learning-Free Heterogeneous LiDAR Place Recognition via Intermediate Mixed-Radix Structural Keypoint Tokenization](https://mlsanity.com/paper/2607.16862/) (2026)

## Abstract (arXiv)

Typical LiDAR SLAM architectures feature a front-end for odometry estimation and a back-end for refining and optimizing the trajectory and map, commonly through loop closures. However, loop closure detection in large-scale missions presents significant computational challenges due to the need to identify, verify, and process numerous candidate pairs for pose graph optimization. Keyframe sampling bridges the front-end and back-end by selecting frames for storing and processing during global optimization. This article proposes an online keyframe sampling approach that constructs the pose graph using the most impactful keyframes for loop closure. We introduce the Minimal Subset Approach (MSA), which optimizes two key objectives: redundancy minimization and information preservation, implemented within a sliding window framework. By operating in the feature space rather than 3-D space, MSA efficiently reduces redundant keyframes while retaining essential information. Evaluations on diverse public datasets show that the proposed approach outperforms naive methods in reducing false positive rates in place recognition, while delivering superior ATE and RPE in metric localization, without the need for manual parameter tuning. Additionally, MSA demonstrates efficiency and scalability by reducing memory usage and computational overhead during loop closure detection and pose graph optimization.