five

so101_pick_ball_vision_only_v9

收藏
Hugging Face2026-06-10 更新2026-06-10 收录
下载链接:
https://huggingface.co/datasets/ofekhamama/so101_pick_ball_vision_only_v9
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是使用LeRobot创建的机器人数据集,专门用于so_follower类型的机器人任务。数据集包含1个完整的情节(episode),总计319帧,涉及1个任务。数据以parquet文件格式存储,总数据文件大小为100MB,视频文件大小为200MB,帧率为15fps。数据集的特征包括:动作(action,6维浮点数,表示机器人关节位置如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置)、观察状态(observation.state,与动作相同的6维浮点数)、两个摄像头图像(observation.images.top和observation.images.wrist,均为480x640分辨率、3通道的RGB视频,使用av1编解码器)、时间戳(timestamp)、帧索引(frame_index)、情节索引(episode_index)、索引(index)和任务索引(task_index)。数据集仅包含训练分割(train split),适用于机器人控制和模仿学习等应用。

This dataset is a robotics dataset created using LeRobot, specifically designed for so_follower-type robotic tasks. It contains 1 complete episode, totaling 319 frames, and involves 1 single task. The data is stored in Parquet file format, with a total data file size of 100 MB, and the total size of the video files is 200 MB, with a frame rate of 15 fps. The dataset's features are as follows: Action (action, 6-dimensional floating-point numbers representing robotic joint positions including shoulder translation, shoulder lift, elbow flexion, wrist flexion, wrist rotation and gripper position), observation.state (6-dimensional floating-point numbers identical to the action), two camera images: observation.images.top and observation.images.wrist, both are 480x640 resolution, 3-channel RGB videos encoded with the AV1 codec, as well as timestamp, frame index, episode index, index and task index. This dataset only includes the training split, and is applicable for applications such as robotic control and imitation learning.
提供机构:
ofekhamama
创建时间:
2026-06-10
原始信息汇总

数据集概述:so101_pick_ball_vision_only_v9

  • 许可证:Apache-2.0
  • 任务类别:机器人学(robotics)
  • 标签:LeRobot

数据集结构

属性
代码库版本 v3.0
机器人类型 so_follower
总回合数 1
总帧数 319
总任务数 1
视频帧率 15 FPS
数据文件大小 100 MB
视频文件大小 200 MB
数据划分 训练集:0:1(全部数据用于训练)

数据特征

特征名称 数据类型 形状/维度 说明
action float32 [6] 机器人动作指令,包含:shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos
observation.state float32 [6] 机器人观测状态,与动作维度相同
observation.images.top video [480, 640, 3] 顶部摄像头视频,AV1编码,YUV420P像素格式,15 FPS
observation.images.wrist video [480, 640, 3] 腕部摄像头视频,AV1编码,YUV420P像素格式,15 FPS
timestamp float32 [1] 时间戳
frame_index int64 [1] 帧索引
episode_index int64 [1] 回合索引
index int64 [1] 索引
task_index int64 [1] 任务索引

数据路径

  • 数据文件data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet
  • 视频文件videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4
二维码
社区交流群
二维码
科研交流群
商业服务