gsethia08/part3-insert_remove_bagging

该数据集是一个用于机器人学研究的双手操作数据集，使用LeRobot框架创建。数据集包含1068个片段，总计284,654帧，覆盖18个不同的任务。数据以30帧每秒（fps）的速率采集，存储为Parquet格式文件，总数据文件大小为100 MB，视频文件大小为200 MB。数据集的主要特征包括：1) 观测数据：包含来自第一人称视角的相机图像（分辨率1080x1920，RGB格式）和双手的状态信息（如手腕和手指的位置和旋转）；2) 动作数据：48维的浮点数向量，表示机器人的动作；3) 相机参数：包括相机的外参和内参；4) 语言指令：字符串类型的指令；5) 时间戳和索引信息：如帧索引、片段索引等。数据集适用于训练和评估机器人操作模型，特别是涉及双手协调的任务。This dataset is a bimanual manipulation dataset for robotics research, created using the LeRobot framework. It contains 1068 episodes, totaling 284,654 frames, and covers 18 distinct tasks. The data is collected at 30 frames per second (fps), stored in Parquet format files, with a total data file size of 100 MB and video file size of 200 MB. Key features of the dataset include: 1) Observations: comprising camera images from a first-person perspective (resolution 1080x1920, RGB format) and state information for both hands (such as wrist and finger positions and rotations); 2) Actions: 48-dimensional float vectors representing robot actions; 3) Camera parameters: including extrinsic and intrinsic camera parameters; 4) Language instructions: string-type instructions; 5) Timestamp and index information: such as frame index, episode index, etc. The dataset is suitable for training and evaluating robotic manipulation models, particularly for tasks involving bimanual coordination.