GAIPAT
收藏资源简介:
GAIPAT数据集由法国格勒诺布尔大学等机构创建,旨在研究人类动作与视线在共享工作环境中的耦合关系,特别是与协作机器人的交互。该数据集包含约80名参与者在模拟工业组装任务中的视线和动作数据。数据通过两种不同的眼动追踪设备(头戴式和远程)收集,参与者在坐立两种姿势下完成操作。数据集可应用于人类意图预测、生成性视线模型以及增强人机协作等领域,但任务相对简单,且仅使用红外眼动追踪技术,存在一定局限性。
The GAIPAT dataset was developed by institutions including the University of Grenoble Alpes (France) and other relevant organizations, aiming to investigate the coupling relationship between human actions and gaze in shared work environments, particularly interactions with collaborative robots. This dataset contains gaze and motion data from approximately 80 participants who completed simulated industrial assembly tasks. The data was collected via two different eye-tracking devices: head-mounted and remote, with participants performing the tasks in both sitting and standing postures. The GAIPAT dataset has applications in multiple domains including human intention prediction, generative gaze modeling, and enhanced human-robot collaboration. Nevertheless, it possesses certain limitations: the experimental tasks are relatively simple, and only infrared eye-tracking technology was employed during data collection.

- 1GAIPAT -Dataset on Human Gaze and Actions for Intent Prediction in Assembly Tasks法国格勒诺布尔大学, CNRS, Grenoble INP, LIG · 2025年



