CoRBS: Comprehensive RGB-D Benchmark for SLAM using Kinect v2

KinectV2 MotionCapture 3Dscanner trajectory map

On this website we provide a new Comprehensive RGB-D Benchmark for SLAM (CoRBS). In contrast to state-of-the-art RGB-D SLAM benchmarks, we provide the combination of real depth and color data together with a ground truth trajectory of the camera and a ground truth 3D model of the scene. Our novel benchmark allows for the first time to independently evaluate the localization as well as the mapping part of RGB-D SLAM systems with real data. We obtained the ground truth for the trajectory using an external motion capture system and for the scene geometry via an external 3D scanner, each with sub-millimeter precision. With precise calibration and systematic validation we ensured the high quality of CoRBS. Our dataset contains twenty image sequences of four different scenes captured with a Kinect v2. We provide all data in a global coordinate system to enable direct evaluation without any further alignment or calibration.
Paper: [pdf]
Contact: Oliver Wasenmüller [link]

Different Scenes:

Human Desk Electrical Racing Car

Raw and Pre-Registered Data:
Since the color and depth images are not registered a priori, we decided to provide two kinds of image sequence per dataset. The first kind are the original unregistered full resolution depth (512×424) and color (1920×1080) images. With these images one can apply an own calibration. The second kind are registered depth and color images, where we applied the transformation as described in the paper. In order to make the images easy applicable in many existing implementations, we provide both the color and the depth in a resolution of 640×480. For more details on the calibration look here.

We are happy to share our data with other researchers. If you use this dataset for scientific publications, please refer to our publication as listed below:

title={{CoRBS}: Comprehensive RGB-D Benchmark for SLAM using Kinect v2},
author={Wasenm\”uller, Oliver and Meyer, Marcel and Stricker, Didier},
booktitle={IEEE Winter Conference on Applications of Computer Vision (WACV)},

All data of CoRBS has been released under Creative Commons 3.0 Attribution License (CC-BY 3.0).


This work was partially funded by the Federal Ministry of Education and Research (Germany) in the context of the projects ARVIDA and Body Analyzer. We thank Johannes Köhler and Tobias Nöll from 3Digify for providing their 3D scanner and the fruitful discussions. Furthermore, we want to thank the racing team Karat Kaiserslautern for providing their car. We also thank Santosh Shah for creating this website.

This website is hosted by DFKI. Retrieve further details here.