HAGS: Hand and Glove Segmentation Dataset
收藏资源简介:
HAGS数据集由德克萨斯大学奥斯汀分校机械工程系创建,专注于工业手套箱环境中的人机协作手和手套分割。该数据集包含1728帧的像素级标签,涵盖191个视频,总计9小时的内容。数据集的创建过程包括收集12名参与者的视频,设计了Jenga塔和工具箱拆解任务,以及使用绿色屏幕和手套穿戴情况来增加数据多样性。HAGS数据集主要用于评估和开发在工业环境中的人机协作安全系统,特别是在需要实时手部分割的应用中。
The HAGS dataset was created by the Department of Mechanical Engineering at The University of Texas at Austin, focusing on human-robot collaborative hand and glove segmentation in industrial glovebox environments. This dataset contains pixel-level annotations for 1,728 frames, spanning 191 videos with a total duration of 9 hours. The dataset creation process involved collecting videos from 12 participants, designing Jenga tower and toolbox disassembly tasks, and utilizing green screens and varied glove-wearing scenarios to enhance data diversity. The HAGS dataset is primarily used for evaluating and developing human-robot collaborative safety systems in industrial settings, particularly for applications requiring real-time hand segmentation.




