PETRAW(PEg TRAnsfer Workflow recognition by different modalities)
收藏OpenDataLab2026-05-24 更新2024-05-09 收录
下载链接:
https://opendatalab.org.cn/OpenDataLab/PETRAW
下载链接
链接失效反馈官方服务:
资源简介:
PETRAW 数据集由 150 个钉转移训练课程序列组成。挂钩转移会话的目标是从左到右再转移 6 个块。每个块必须用一只手从钉子中取出,转移到另一只手,然后插入板另一侧的钉子中。所有病例均由雷恩大学 LTSI 实验室的非医学专家获得。数据集分为90个案例组成的训练数据集和60个案例组成的测试数据集。一个案例由运动学数据、视频、每帧的语义分割和工作流注释组成。
The PETRAW dataset consists of 150 sequences of peg transfer training tasks. The goal of each peg transfer session is to transfer 6 blocks from left to right. Each block must be picked from a peg using one hand, transferred to the other hand, and then inserted into a peg on the opposite side of the board. All sessions were collected by non-medical experts from the LTSI Laboratory of the University of Rennes. The dataset is split into a training set containing 90 cases and a test set with 60 cases. Each case includes kinematic data, video footage, per-frame semantic segmentation annotations, and workflow annotations.
提供机构:
OpenDataLab
创建时间:
2022-08-16
搜集汇总
数据集介绍

背景与挑战
背景概述
PETRAW 数据集包含150个钉转移训练课程序列,旨在从左到右转移6个块,每个案例整合了运动学数据、视频、语义分割和工作流注释。数据由非医学专家采集,分为90个训练案例和60个测试案例,用于多模态工作流识别研究。
以上内容由遇见数据集搜集并总结生成



