ROCVOct 20, 2016

ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras

arXiv:1610.06475v26342 citations
Originality Highly original
AI Analysis

This provides an open-source, real-time SLAM system for researchers and practitioners in robotics and computer vision, enabling accurate localization and mapping across diverse environments.

The authors tackled the problem of simultaneous localization and mapping (SLAM) for various camera types by developing ORB-SLAM2, a complete system that achieves state-of-the-art accuracy on 29 public sequences, often being the most accurate solution.

We present ORB-SLAM2 a complete SLAM system for monocular, stereo and RGB-D cameras, including map reuse, loop closing and relocalization capabilities. The system works in real-time on standard CPUs in a wide variety of environments from small hand-held indoors sequences, to drones flying in industrial environments and cars driving around a city. Our back-end based on bundle adjustment with monocular and stereo observations allows for accurate trajectory estimation with metric scale. Our system includes a lightweight localization mode that leverages visual odometry tracks for unmapped regions and matches to map points that allow for zero-drift localization. The evaluation on 29 popular public sequences shows that our method achieves state-of-the-art accuracy, being in most cases the most accurate SLAM solution. We publish the source code, not only for the benefit of the SLAM community, but with the aim of being an out-of-the-box SLAM solution for researchers in other fields.

Code Implementations7 repos

Data from Papers with Code (CC-BY-SA-4.0)

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes