[{"@type":"PropertyValue","name":"Data content","value":"135,000 hours of multi-scene ego-centric data, each set contains: ①raw video, ②IMU sensor data, ③Slam and depth data, ④hand keypoints, ⑤subtask annotation"},{"@type":"PropertyValue","name":"Data distribution","value":"covers diverse scenes and tasks including household operations, office light tasks, supermarket/retail, and industrial/semi-structured operations."},{"@type":"PropertyValue","name":"Data quality","value":"①complete data without omission: video frames, IMU, depth maps, hand keypoints, and annotations are matched; ②cross-modal timestamp alignment without temporal offset; ③unified data formats with compliant data types."},{"@type":"PropertyValue","name":"Data formats","value":"①raw video: mp4, ②IMU sensor data: json, ③Slam and depth data: json, mp4, ④hand keypoints: json, ⑤subtask annotations: json"}]
{"id":2204,"datatype":"1","titleimg":"https://www.nexdata.ai/shujutang/static/image/index/datatang_tuxiang_default.webp","type1":"282","type1str":null,"type2":"283","type2str":null,"dataname":"135,000 Hours - Ego-centric Data_D","datazy":[{"title":"Data content","content":"135,000 hours of multi-scene ego-centric data, each set contains: ①raw video, ②IMU sensor data, ③Slam and depth data, ④hand keypoints, ⑤subtask annotation"},{"title":"Data distribution","content":"covers diverse scenes and tasks including household operations, office light tasks, supermarket/retail, and industrial/semi-structured operations."},{"title":"Data quality","content":"①complete data without omission: video frames, IMU, depth maps, hand keypoints, and annotations are matched; ②cross-modal timestamp alignment without temporal offset; ③unified data formats with compliant data types."},{"title":"Data formats","content":"①raw video: mp4, ②IMU sensor data: json, ③Slam and depth data: json, mp4, ④hand keypoints: json, ⑤subtask annotations: json"}],"datatag":"ego-centric,slam,depth,handpose,subtask","technologydoc":null,"downurl":null,"datainfo":null,"standard":null,"dataylurl":null,"flag":null,"publishtime":null,"createby":null,"createtime":null,"ext1":null,"samplestoreloc":null,"hosturl":null,"datasize":null,"industryPlan":null,"keyInformation":null,"samplePresentation":[{"name":"GX016878_visualized_clipped.mp4","url":"https://storage-product.datatang.com/damp/product/sample_presentation/20261103155719/GX016878_visualized_clipped.mp4?Expires=4102415999&OSSAccessKeyId=LTAI5tEBeSWUJiqjXvBMsxEu&Signature=Y%2BU8wVEIgSYuPCdMX9kfvFxMhlI%3D","intro":"","size":38387019,"progress":100,"type":"mp4"},{"name":"GX016920_visualized_clipped.mp4","url":"https://storage-product.datatang.com/damp/product/sample_presentation/20261103155719/GX016920_visualized_clipped.mp4?Expires=4102415999&OSSAccessKeyId=LTAI5tEBeSWUJiqjXvBMsxEu&Signature=vVFQI7GSLchQfw9rrH4NjDPOXV8%3D","intro":"","size":28710285,"progress":100,"type":"mp4"},{"name":"GX016989_visualized_clipped.mp4","url":"https://storage-product.datatang.com/damp/product/sample_presentation/20261103155719/GX016989_visualized_clipped.mp4?Expires=4102415999&OSSAccessKeyId=LTAI5tEBeSWUJiqjXvBMsxEu&Signature=wmRjj9cC%2FlVufBz4Wg2P%2B3CHSTI%3D","intro":"","size":25198700,"progress":100,"type":"mp4"}],"officialSummary":"135,000 Hours - Ego-centric Data_D, each set contains ①raw video, ②IMU sensor data, ③Slam and depth data, ④hand keypoints, ⑤subtask annotation, covers diverse scenes and tasks including household operations, office light tasks, supermarket/retail, and industrial/semi-structured operations. This dataset can be used for tasks like embodied AI model training.","dataexampl":null,"datakeyword":["ego-centric","slam","depth","handpose","subtask"],"isDelete":null,"ids":null,"idsList":null,"datasetCode":null,"productStatus":null,"tagTypeEn":"Type","tagTypeZh":null,"website":null,"samplePresentationList":null,"datazyList":null,"keyInformationList":null,"dataexamplList":null,"bgimg":null,"datazyScriptList":null,"datakeywordListString":null,"sourceShowPage":"embodiedAi","dataShowType":"[{\"code\":\"0\",\"language\":\"ZH\"},{\"code\":\"1\",\"language\":\"ZH\"},{\"code\":\"2\",\"language\":\"EN,JP,KO\"},{\"code\":\"3\",\"language\":\"EN\"}]","productNameEn":"135,000 Hours - Ego-centric Data_D","BGimg":"","voiceBg":["/shujutang/static/image/comm/audio_bg.webp","/shujutang/static/image/comm/audio_bg2.webp","/shujutang/static/image/comm/audio_bg3.webp","/shujutang/static/image/comm/audio_bg4.webp","/shujutang/static/image/comm/audio_bg5.webp"]}
135,000 Hours - Ego-centric Data_D, each set contains ①raw video, ②IMU sensor data, ③Slam and depth data, ④hand keypoints, ⑤subtask annotation, covers diverse scenes and tasks including household operations, office light tasks, supermarket/retail, and industrial/semi-structured operations. This dataset can be used for tasks like embodied AI model training.
This is a paid dataset licensed for commercial use. Ready-made datasets are available for immediate integration into AI projects.
Specifications
Data content
135,000 hours of multi-scene ego-centric data, each set contains: ①raw video, ②IMU sensor data, ③Slam and depth data, ④hand keypoints, ⑤subtask annotation
Data distribution
covers diverse scenes and tasks including household operations, office light tasks, supermarket/retail, and industrial/semi-structured operations.
Data quality
①complete data without omission: video frames, IMU, depth maps, hand keypoints, and annotations are matched; ②cross-modal timestamp alignment without temporal offset; ③unified data formats with compliant data types.
What types of Embodied AI applications can Nexdata’s datasets support?
Nexdata’s Embodied AI datasets are designed to support a wide range of robotics and embodied intelligence applications, including robot perception, manipulation, navigation, human-robot interaction, and Vision-Language-Action (VLA) model development. Depending on the dataset, data may include multimodal video, sensor data, robot trajectories, actions, poses, and other annotations for training and evaluation.
Can Nexdata customize Embodied AI datasets based on our specific requirements?
Yes. If our off-the-shelf datasets do not fully meet your requirements, Nexdata provides flexible custom data collection, annotation, and curation services. We can customize data based on your target robot platform, tasks, environments, sensors, data volume, annotation requirements, and model specifications, helping you build datasets tailored to your specific Embodied AI or VLA project.
Can Nexdata support large-scale Embodied AI data collection projects?
Yes. Nexdata operates an 8,000-square-meter real-world data collection dojo that enables large-scale and diverse data collection for robotics and Embodied AI applications. The facility can support customized environments, task scenarios, robot operations, and multimodal data collection, allowing us to accommodate projects with complex requirements and large data volumes.
Specifications
Sample
Recommended Dataset
