F-SIOL-310
收藏arXiv2022-04-21 更新2024-06-21 收录
下载链接：
https://tinyurl.com/yb38syd5
下载链接
链接失效反馈官方服务：
资源简介：
F-SIOL-310是一个专为机器人视觉设计的少样本增量对象学习数据集，由宾夕法尼亚州立大学创建。该数据集包含22个类别的310个常见家用物品，每个类别有11-17个不同的物品。数据集通过Baxter机器人自主捕获，考虑了多种机器人视觉挑战，如物体大小、透明度等。F-SIOL-310特别适用于测试少样本增量学习能力，旨在解决机器人如何在有限的人类协助下，通过少数示例快速学习新对象的问题。
F-SIOL-310 is a few-shot incremental object learning dataset specifically designed for robotic vision, created by Pennsylvania State University. This dataset includes 310 common household items across 22 categories, with 11 to 17 distinct items per category. The dataset was autonomously captured by the Baxter robot, taking into account various robotic vision challenges such as object size and transparency. F-SIOL-310 is particularly suitable for testing few-shot incremental learning capabilities, and aims to solve the problem of how robots can quickly learn new objects from a small number of examples with limited human assistance.
提供机构：
宾夕法尼亚州立大学
创建时间：
2021-03-23
