An evaluation of the RGB-D SLAM system

TLDR

The authors propose an RGB‑D SLAM system for hand‑held Microsoft Kinect cameras and evaluate its performance on a diverse indoor dataset. The system simultaneously estimates the Kinect trajectory and builds a dense 3D model, assessing accuracy, robustness, and runtime using SIFT, SURF, and ORB feature descriptors. Experiments show the system robustly handles challenging indoor data, operates online, and is released as open‑source.

Abstract

We present an approach to simultaneous localization and mapping (SLAM) for RGB-D cameras like the Microsoft Kinect. Our system concurrently estimates the trajectory of a hand-held Kinect and generates a dense 3D model of the environment. We present the key features of our approach and evaluate its performance thoroughly on a recently published dataset, including a large set of sequences of different scenes with varying camera speeds and illumination conditions. In particular, we evaluate the accuracy, robustness, and processing time for three different feature descriptors (SIFT, SURF, and ORB). The experiments demonstrate that our system can robustly deal with difficult data in common indoor scenarios while being fast enough for online operation. Our system is fully available as open-source.