Projects per year
Description
Subject: Manufacturing Engineering + Human-Robot Interaction Specific Subject Area: 3D point-cloud data of human movements and gestures captured by a RaDAR sensor synchronized with RGB-D images installed on the base of a robot manipulator. Type of Data: 3D point-cloud, 3D skeleton keypoints; Raw, Labelled Data Collection: 3D point-clouds of humans were collected using an IWR6843AOPEVM RaDAR at a 10 Hz data frequency. 3D keypoints were extracted by the Nuitrack SDK from RGB-D images captured by an Intel RealSense L515 camera at 30 frames per second as the ground truth. Both sensors' frames were synchronized. An offset (3.9 cm in both horizontal and vertical directions) and a 90-degree rotation around the local x-axis were applied to the camera data for spatial calibration and alignment with the RaDAR's coordinate system. Data Source Location: Brubotics, Vrije Universiteit Brussels, Elsene, Brussels, Belgium Related Research Article: Not yet available (Under Submission)
format
CSV
Date made available | 15 Aug 2024 |
---|---|
Publisher | Mendeley Data |
Keywords
- Gesture Recognition
- Radar
- Skeleton
- Safety
- Robot
- Robotics
Format
- Format
- csv
Projects
- 1 Finished
Research output
- 1 Article
-
mmPrivPose3D: A dataset for pose estimation and gesture command recognition in human-robot collaboration using frequency modulated continuous wave 60Hhz RaDAR
Roshandel, N., Scholz, C., Cao, H-L., Amighi, M., Firouzipouyaei, H., Burkiewicz, A., Menet, S., Ballen-Moreno, F., Sisavath, D. W., Imrith, E., Paolillo, A., Genoe, J. & Vanderborght, B., Apr 2025, In: Data in Brief. 59, p. 1-8 8 p., 111316.Research output: Contribution to journal › Article › peer-review
Open AccessFile1 Downloads (Pure)