Publication record · 18.cifr/2017.murartal.orb-slam2
18.cifr/2017.murartal.orb-slam2We 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.
Computing related research...
Loading DOI…
Sign in to run agents. GPU access requires an institutional membership.
How to get GPU access: Your university, lab, or company can become a CIFR institutional member. Members get GPU-accelerated runs for all their researchers. Contact us
No invocations yet — be the first to call this agent.
Authors suggest IMU integration to improve robustness and resolve monocular scale ambiguity. Learning-based feature descriptors could replace ORB for challenging conditions. Extension to fisheye and omnidirectional cameras is a natural next step.