CADDY Underwater Stereo-Vision Dataset for Human–Robot Interaction (HRI) in the Context of Diver Activities

Journal of Marine Science and Engineering

卷:7

Andrea Ranieri¹  Andreas Birk²  Enrica Zereik³  Davide Chiarella⁴  Anja Babić⁴  Arturo Gomez Chavez⁵ 

[1] Electrical Engineering, Jacobs University Bremen, Campus Ring 1, 28759 Bremen, Germany;

[2] Faculty of Electrical Engineering and Computing, University of Zagreb, Unska ul. 3, 10000 Zagreb, Croatia;

[3] Institute for Computational Linguistics—National Research Council, Via E. De Marini 6, 16149 Genova, Italy;

[4] Institute of Marine Engineering—National Research Council, Via E. De Marini 6, 16149 Genova, Italy;

[5] Robotics Group, Computer Science &

关键词: dataset;    underwater imaging;    image processing;    marine robotics;    field robotics;    human–robot interaction;    stereo vision;    object classification;    human pose estimation;   

DOI  :  10.3390/jmse7010016

来源: DOAJ

【 摘 要 】

In this article, we present a novel underwater dataset collected from several field trials within the EU FP7 project “Cognitive autonomous diving buddy (CADDY)”, where an Autonomous Underwater Vehicle (AUV) was used to interact with divers and monitor their activities. To our knowledge, this is one of the first efforts to collect a large public dataset in underwater environments with the purpose of studying and boosting object classification, segmentation and human pose estimation tasks. The first part of the dataset contains stereo camera recordings (≈10 K) of divers performing hand gestures to communicate with an AUV in different environmental conditions. The gestures can be used to test the robustness of visual detection and classification algorithms in underwater conditions, e.g., under color attenuation and light backscatter. The second part includes stereo footage (≈12.7 K) of divers free-swimming in front of the AUV, along with synchronized measurements from Inertial Measurement Units (IMU) located throughout the diver’s suit (DiverNet), which serve as ground-truth for human pose and tracking methods. In both cases, these rectified images allow the investigation of 3D representation and reasoning pipelines from low-texture targets commonly present in underwater scenarios. This work describes the recording platform, sensor calibration procedure plus the data format and the software utilities provided to use the dataset.

【 授权许可】

Unknown