# cuVSLAM: CUDA accelerated visual odometry and mapping

arXiv: https://arxiv.org/abs/2506.04359 · HTML: https://mlsanity.com/paper/2506.04359/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 24 references that are in the map.
- First posted on arXiv 4 Jun 2025.

## Builds on (most cited first, 8 of 8)

- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [The TUM VI Benchmark for Evaluating Visual-Inertial Odometry](https://mlsanity.com/paper/1804.06120/) (2018)
- [On-Manifold Preintegration for Real-Time Visual-Inertial Odometry](https://mlsanity.com/paper/1512.02363/) (2015)
- [The Marathon 2: A Navigation System](https://mlsanity.com/paper/2003.00368/) (2020)
- [TartanGround: A Large-Scale Dataset for Ground Robot Perception and Navigation](https://mlsanity.com/paper/2505.10696/) (2025)
- [M2DGR: A Multi-sensor and Multi-scenario SLAM Dataset for Ground Robots](https://mlsanity.com/paper/2112.13659/) (2021)
- [nvblox: GPU-Accelerated Incremental Signed Distance Field Mapping](https://mlsanity.com/paper/2311.00626/) (2023)
- [Hilti SLAM Challenge 2023: Benchmarking Single + Multi-session SLAM across Sensor Constellations in Construction](https://mlsanity.com/paper/2404.09765/) (2024)

## Built on by (most cited first, 4 of 4)

- [Offload or Overload: A Platform Measurement Study of Mobile Robotic Manipulation Workloads](https://mlsanity.com/paper/2603.18284/) (2026)
- [DGSG-Mind: Dynamic 3D Gaussian Scene Graphs for Long-Term Scene Understanding and Grounding](https://mlsanity.com/paper/2605.29879/) (2026)
- [MAVP: Map-Aware Visuomotor Policies for Mobile Manipulation](https://mlsanity.com/paper/2609.26378/) (2026)
- [RoboCap: A New Platform for Egocentric Robot Learning](https://mlsanity.com/paper/2610.07217/) (2026)

## Abstract (arXiv)

Accurate and robust pose estimation is a key requirement for any autonomous robot. We present cuVSLAM, a state-of-the-art solution for visual simultaneous localization and mapping, which can operate with a variety of visual-inertial sensor suites, including multiple RGB and depth cameras, and inertial measurement units. cuVSLAM supports operation with as few as one RGB camera to as many as 32 cameras, in arbitrary geometric configurations, thus supporting a wide range of robotic setups. cuVSLAM is specifically optimized using CUDA to deploy in real-time applications with minimal computational overhead on edge-computing devices such as the NVIDIA Jetson. We present the design and implementation of cuVSLAM, example use cases, and empirical results on several state-of-the-art benchmarks demonstrating the best-in-class performance of cuVSLAM.