xiaochyVera/insert_donut_human_3

该数据集是由LeRobot创建的机器人相关数据集，包含200个训练集片段，总计56652帧数据。数据集记录了机器人关节状态（8个关节）、末端执行器姿态（6D旋转+3D平移+夹持器开合）、来自Azure Kinect左前摄像头的深度图像和彩色图像（720x1280分辨率）等多种传感器数据。数据以30fps的帧率采集，存储为parquet格式，并包含视频文件。

This dataset is a robotics-related dataset created by LeRobot, containing 200 training episodes with a total of 56,652 frames. The dataset records various sensor data including robot joint states (8 joints), end-effector poses (6D rotation + 3D translation + gripper articulation), depth and color images (720x1280 resolution) from Azure Kinect left and front cameras. Data was collected at 30fps and stored in parquet format, along with video files.