cuVSLAM: CUDA accelerated visual odometry and mapping

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Korovko, Alexander, Slepichev, Dmitry, Efitorov, Alexander, Dzhumamuratova, Aigul, Kuznetsov, Viktor, Rabeti, Hesam, Biswas, Joydeep, Pouya, Soha
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918085811765248
author Korovko, Alexander
Slepichev, Dmitry
Efitorov, Alexander
Dzhumamuratova, Aigul
Kuznetsov, Viktor
Rabeti, Hesam
Biswas, Joydeep
Pouya, Soha
author_facet Korovko, Alexander
Slepichev, Dmitry
Efitorov, Alexander
Dzhumamuratova, Aigul
Kuznetsov, Viktor
Rabeti, Hesam
Biswas, Joydeep
Pouya, Soha
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04359
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle cuVSLAM: CUDA accelerated visual odometry and mapping
Korovko, Alexander
Slepichev, Dmitry
Efitorov, Alexander
Dzhumamuratova, Aigul
Kuznetsov, Viktor
Rabeti, Hesam
Biswas, Joydeep
Pouya, Soha
Robotics
Artificial Intelligence
Software Engineering
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.
title cuVSLAM: CUDA accelerated visual odometry and mapping
topic Robotics
Artificial Intelligence
Software Engineering
url https://arxiv.org/abs/2506.04359