Effects of Object Characteristics on Manipulations

The dataset is designed for investigating the impact of object-specific attributes on human pick-and-place movements, while also enabling comparison of motion kinematics captured by diverse sensor modalities.
This dataset is intended to facilitate discussions surrounding broader learning challenges in the domain of hand-object interaction, including intention recognition and motion generation, with direct applications in Robotics research.
It comprises recordings from 15 subjects executing 80 repetitions of a pick-and-place task under various experimental conditions, resulting in a total of 1200 pick-and-place instances.
Data collection was facilitated by a multimodal setup incorporating multiple cameras for varied perspectives, a motion capture system, and a wrist-worn inertial measurement unit. The objects manipulated within the experiments exhibit identical shape, size, and appearance but vary in weight and liquid filling, thus impacting the requisite level of delicacy during handling.

The detailed description is available as a dataset paper:
Lastrico L, Belcamino V, Carfì A, et al. ***The effects of selected object features on a pick-and-place task: A human multimodal dataset*. The International Journal of Robotics Research**. 2024;43(1):98-109. [doi:10.1177/02783649231210965](https://doi.org/10.1177/02783649231210965)

Software helper functions can be accessed from [Github](https://github.com/lindalastrico/objectsManipulationDataset).

本数据集旨在探讨特定物体属性对人类拾取与放置动作的影响，并允许比较由不同传感器模式捕捉的运动学。该数据集旨在促进围绕手与物体交互领域更广泛学习挑战的讨论，包括意图识别和运动生成，其直接应用于机器人研究。数据集包含15名受试者在不同实验条件下执行80次拾取与放置任务的记录，总计生成1200个拾取与放置实例。数据收集得益于一种多模态设置，该设置集成了多台摄像机以提供不同的视角、运动捕捉系统和佩戴在手腕上的惯性测量单元。实验中所操纵的物体在形状、大小和外观上均相同，但重量和液体填充量不同，从而影响了处理过程中所需的精细程度。详细描述可在数据集论文中找到：Lastrico L, Belcamino V, Carfì A, et al. **《选定物体特征对拾取与放置任务的影响：一项人类多模态数据集》**. 国际机器人研究杂志. 2024;43(1):98-109. [doi:10.1177/02783649231210965](https://doi.org/10.1177/02783649231210965)。辅助软件功能可通过[GitHub](https://github.com/lindalastrico/objectsManipulationDataset)访问。