Context-Aware Activity Recognition in Logistics (CAARL) – A optical marker-based Motion Capture Dataset

CAARL is a freely accessible logistics-dataset for human activity recognition, which contains human movement and context information from two subjects. The context information includes the positions of objects such as two picking carts, a packaging table, different racks, a base and three entrances. In the ’Innovationlab Hybrid Services in Logistics’ at TU Dortmund University, two picking and one packing scenarios were recorded using an optical marker based motion capture system. Each subject and object is equipped with several markers. 140 minutes of human movements have been labelled and categorised into 8 activity classes and 19 binary coarse-semantic descriptions, also called attributes. The labelled human movements are synchronised with the context information. They have exactly the same sampling rate (same start and end). The oMoCap data is in csv format. Further formats (e.g. C3D) are available on request. CAARL is based on the set-up and scenarios of the LARa dataset, which contains only human movements. Information about LARa can be found in the dataset and the associated paper: Dataset: “Logistic Activity Recognition Challenge (LARa) – A Motion Capture and Inertial Measurement Dataset”, Zenodo 2020, DOI: 10.5281/zenodo.3862782 Paper: “LARa: Creating a Dataset for Human Activity Recognition in Logistics Using Semantic Attributes”, Sensors 2020, DOI: 10.3390/s20154083 If you use the CAARL dataset for research, please cite the following paper: “Context-Aware Human Activity Recognition in Industrial Processes”, Sensors 2021, DOI: 10.3390/s22010134

CAARL是一款可免费获取的、面向人类活动识别的物流数据集，涵盖两名受试者的人体运动数据与上下文信息。其中上下文信息包含两台拣选推车、一张包装工作台、不同规格货架、一处基座以及三处入口等物体的位置数据。
本数据集在多特蒙德工业大学（TU Dortmund University）的“物流混合服务创新实验室”中完成采集，采用基于光学标记的运动捕捉系统 (optical marker based motion capture system)，记录了两类拣选作业与一类包装作业场景。每位受试者与被追踪物体均配备了多个标记点。
研究人员已对总计140分钟的人体运动数据完成标注，将其划分为8类活动类别与19项二元粗语义描述（亦称为属性）。标注后的人体运动数据与上下文信息已完成同步，二者采样率完全一致，且起始、结束时间完全匹配。
光学运动捕捉 (optical motion capture，简称oMoCap) 数据采用CSV格式存储，其他格式（如C3D）可按需申请获取。
CAARL数据集基于仅包含人体运动数据的LARa (Logistic Activity Recognition Challenge) 数据集的采集设置与场景构建。关于LARa数据集的详细信息可参考其配套数据集与学术论文：
数据集：《Logistic Activity Recognition Challenge (LARa) – 一款运动捕捉与惯性测量数据集》，Zenodo 2020，DOI: 10.5281/zenodo.3862782
论文：《LARa：基于语义属性构建物流领域人类活动识别数据集》，Sensors 2020，DOI: 10.3390/s20154083
若您在研究工作中使用CAARL数据集，请引用如下学术论文：《工业场景下的上下文感知人类活动识别》，Sensors 2021，DOI: 10.3390/s22010134

搜集汇总

背景与挑战

背景概述

CAARL是一个专注于物流领域的人类活动识别数据集，它基于光学标记运动捕捉技术，记录了拣选和包装场景中两个受试者的140分钟运动数据，并包含物体位置等上下文信息。数据被精细标注为8个活动类和19个二进制属性，且运动与上下文信息同步，采样率一致，总大小11.3 GB，主要格式为CSV。该数据集扩展了LARa数据集，增加了上下文维度，适用于物流活动识别研究。