eunjuri/filled_can_img_depth
收藏资源简介:
该数据集使用LeRobot创建,与机器人技术相关,特别是Unitree_G1_Inspire机器人。数据集包含82个片段,总计37,941帧和164个视频。数据包括观察(状态、语言标记、图像、深度)、动作、时间戳和帧索引。观察状态和动作各有26个维度,详细描述了机器人的各个关节状态。此外,还包括语言标记、注意力掩码以及来自左侧高摄像头的视频数据。数据以parquet文件格式存储,视频为mp4格式。
This dataset was created using LeRobot and is related to robotics, specifically the Unitree_G1_Inspire robot. It contains 82 episodes with a total of 37,941 frames and 164 videos. The data includes observations (state, language tokens, images, depths), actions, timestamps, and frame indices. The observation state and action each have 26 dimensions, detailing the state of various robot joints. Additionally, it includes language tokens, attention masks, and video data from a left high camera. The data is stored in parquet files and videos are in mp4 format.




