{"success":true,"data":{"author":"Eidon AI (Solidic Labs Inc)","bucketAccessType":"public","bucketName":"eidon-ai/tracker-pov","bucketPrefix":"","category":"Robotics","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:10.750Z","description":"Eidon Tracker POV is a final release of 1,274 hours of egocentric video recorded while contributors did ordinary household work in their own homes, such as laundry, cleaning, dishes and cooking. Contributors wore a head-mounted camera and a seven-sensor IMU harness, so each recording pairs first-person video with 24 Hz orientation data for both hands, both forearms, both upper arms and the chest. The card reports 13,451 recordings from 27 contributors; video is MP4 from 1080p to 4K, mostly at 30 fps. This repo holds the video under recordings/data/ (in numbered folders of 1,000 files) plus recordings/metadata.parquet, one row per recording with duration, task type, timestamp and contributor id, and users.parquet. The matching IMU streams are in the separate eidon-ai/tracker-pov-imu repo (about 9.5 GB, joined on recording_id). Measured 2026-11-14 at about 9 TB. Read metadata with pandas and video with PyAV, decord or OpenCV. Useful for learning manipulation from human video, activity recognition and motion-to-video modelling. Published under CC BY 4.0, with attribution required.","displayTitle":"Eidon Tracker POV: Egocentric Household Video with IMU Arm Tracking","id":"29410289-00bf-520e-8139-d9e59d1f5741","license":"CC-BY-4.0","licenseEvidenceUrl":"https://huggingface.co/datasets/eidon-ai/tracker-pov","objectCountEstimate":13455,"probeStatus":"ok","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"eidon-ai/tracker-pov\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"eidon-ai/tracker-pov\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=eidon-ai, license:cc-by-4.0, size_categories:10K<n<100K, downloads=59929"],"format":"mixed","generatedAt":"2026-11-14T14:15:19.402Z","layout":{"example":"recordings/metadata.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"recordings/metadata.parquet","truncatedAt":512}],"sampleObjects":["recordings/data/000/1000_video.mp4","recordings/data/000/100_video.mp4","recordings/data/000/101_video.mp4","recordings/data/000/102_video.mp4","recordings/data/000/103_video.mp4"],"sidecars":[{"bytes":"---\nlicense: cc-by-4.0\npretty_name: Eidon Tracker POV\nsize_categories:\n- 10K<n<100K\ntask_categories:\n- robotics\n- video-classification\ntags:\n- egocentric\n- imu\n- manipulation\n- activities-of-daily-living\n- motion-capture\n- embodied-ai\nconfigs:\n- config_name: recordings\n  data_dir: recordings\n  default: true\n  drop_labels: true\nbuckets:\n- eidon-ai/egocentric-pov\n---\n\n# Eidon Tracker POV\n\n**1,274 hours of egocentric video paired with 7-point IMU arm tracking, recorded during ordinary household work.**\n\nContributors wore a head-mounted camera and a seven-sensor IMU harness while doing real chores in their own homes: laundry, cleaning, dishes, cooking. Each recording pairs first-person video with 24 Hz orientation data for both hands, both forearms, both upper arms, and the chest.\n\nThis is a complete, final release. Eidon AI (Solidic Labs Inc) has wound down; the dataset is published under CC-BY-4.0 so it outlives the company.\n\n> ### This repo holds the video. The sensor data is next door.\n>\n> The release sits in three places:\n>\n> | | Contents | Size |\n> |---|---|---|\n> | this repo (`tracker-pov`) | the 13,451 MP4s and `metadata.parquet` | 9.05 TB |\n> | [`tracker-pov-imu`](https://huggingface.co/datasets/eidon-ai/tracker-pov-imu) | the IMU streams for these same recordings, 779M rows | 9.5 GB |\n> | [`egocentric-pov`](https://huggingface.co/buckets/eidon-ai/egocentric-pov) | extra video with no sensor data. A bucket, so `load_dataset` does not reach it | 1.55 TB |\n>\n> The first two are one dataset in two pieces, covering the same 13,451 recordings and joined on\n> `recording_id`. For paired video and motion you want both.\n> The [organization page](https://huggingface.co/eidon-ai) has the overview.\n\n| | |\n|---|---|\n| Recordings | 13,451 |\n| Duration | 1,273.8 hours |\n| Contributors | 27 |\n| Size | 9.05 TB |\n| Video | MP4, 1080p to 4K, mostly 30 fps |\n| Sensors | 7 × 9-axis IMU @ 24 Hz |\n\n## Structure\n\nThe release is split across two repos and a bucket.\n\n| Where | What | Size |\n|---|---|---|\n| **this repo** | 13,451 MP4s + `metadata.parquet` | 9.05 TB |\n| [`eidon-ai/tracker-pov-imu`](https://huggingface.co/datasets/eidon-ai/tracker-pov-imu) | 24 Hz IMU streams, long-form Parquet | ~6.6 GB |\n| [`eidon-ai/egocentric-pov`](https://huggingface.co/buckets/eidon-ai/egocentric-pov) | 306 hrs of video-only footage, no sensors. Storage Bucket, not a dataset | 1.55 TB |\n\n```\n├── README.md\n├── recordings/\n│   ├── metadata.parquet        # one row per recording\n│   └── data/000 … 013/         # 13,451 MP4s, 1,000 per directory\n```\n\nThe `data/NNN/` split exists only to keep directories manageable; the grouping carries no\nmeaning.\n\n```python\nfrom datasets import load_dataset\n\nrecs = load_dataset(\"eidon-ai/tracker-pov\", split=\"train\")\nimu  = load_dataset(\"eidon-ai/tracker-pov-imu\", split=\"train\", streaming=True)\n```\n\nBoth are keyed by `recording_id`. Filter the metadata first, then pull only the IMU you need:\n\n```python\nimport pandas as pd\nmeta = pd.read_parquet(\"hf://datasets/eidon-ai/tracker-pov/recordings/metadata.parquet\")\ncooking = meta[(meta.task_type == \"cooking\") & (meta.qc_status == \"valid\")]\n\nimu = pd.read_parquet(\"hf://datasets/eidon-ai/tracker-pov-imu/imu-0000.parquet\",\n                      filters=[(\"recording_id\", \"in\", cooking.recording_id.tolist())])\n```\n\n## `metadata.parquet`\n\n| Column | Description |\n|---|---|\n| `file_name` | path to the MP4, **relative to `recordings/`** (see note below) |\n| `recording_id` | integer id; joins to the `imu` subset and matches the filename prefix |\n| `contributor_id` | joins to `users.parquet` |\n| `duration_seconds` | recording length |\n| `task_type` | activity label (see below) |\n| `recorded_at` | ISO 8601 timestamp |\n| `sensor_schema_version` | 1 or 2 (see *Raw motion data*) |\n| `qc_status` | `valid`, `flagged`, or `invalid` |\n| `video_width`, `video_height`, `video_fps`, `video_duration_seconds` | from ffprobe |\n| `video_bytes`, `sensor_bytes` | file sizes |\n| `n_slots` | how many of the 7 body slots actually streamed |\n| `has_chest` | whether the chest refere","encoding":"utf8","key":"README.md","name":"README.md","size":11478,"truncated":true}],"sourceObjectKey":"recordings/metadata.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"eidon-ai/tracker-pov\")\nprint(ds)","starterSamplePath":"recordings/data/000/1000_video.mp4","totalObjects":13455},"summary":"Eidon Tracker POV is a final release of 1,274 hours of egocentric video recorded while contributors did ordinary household work in their own homes, such as laundry, cleaning,…","title":"Eidon Tracker POV: Egocentric Household Video with IMU Arm Tracking","updatedAt":"2026-11-14T17:50:31.341Z","visibility":"public","websiteUrl":"https://huggingface.co/eidon-ai"}}