SURP - Surface Processing Dataset

Info The dataset contains videos, motions, and forces of human surface processing demonstrations. Each of the 21 subjects used a powered sanding tool to process 4 different workpieces. Dataset Webpage: https://teiband.github.io/SURP/ Python Code to Read the Dataset: https://github.com/teiband/SURP Conference Paper: https://elib.dlr.de/215557 Reference The conference paper provides a detailed explanation about the data acquistion setup and environment. https://elib.dlr.de/215557 @InProceedings{, author = {Eiband, Thomas and Leimbach, Lars and Nottensteiner, Korbinian and Albu-Schäffer, Alin}, booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, title = {Extraction of Robotic Surface Processing Strategies from Human Demonstrations}, year = {2025}, pages = {(accepted)}, } Files ... per subject folder from "01" to "21": metadata in file meta.yaml video in file [object].mp4 time series data in file [object].pts with the data columns described below Data Columns-------------------------------------------X Y Z Time[s] RX RY RZ RW FX FY FZ TX TY TZ-------------------------------------------X,Y,Z: position of the tracker w.r.t. calibration stationRX, RY, RZ, RW: Quaternion orientation (scalar-last convention) of the tracker w.r.t. calibration stationTime[s]: time in secondsFX, FY, FZ: forces in FTS sensor frameTX, TY, TZ: moments in FTS sensor frame Translation in (x,y,z) from Vive tracker to sanding tool finishing disk center point: [0, 0, 0.2227]