Humanoid robots
Requirements
A humanoid perceives from a head-mounted camera, so custom humanoid dataset capture should be recorded from that same viewpoint: hands entering frame from below, occlusion during grasp, motion during locomotion.
Bimanual coordination and whole-body balance are core to humanoid tasks. Capture needs two wrist cameras plus 3D hand and body pose so both the action and the supporting posture are recorded together.
Ten synchronized streams per episode (4K head camera, two wrist cameras, depth, IMU, 3D hand and body pose, segmentation, and action labels), hardware-aligned under 2 ms.
A perpetual, buyer-owned commercial license backed by a documented consent chain, so a policy trained on the data can go into a product.
Embodiment fit
A humanoid with a head-mounted camera occupies almost exactly the viewpoint a head-worn rig records. Egocentric human capture matches that observation space far more closely than third-person video or single-arm teleoperation, which fix a different camera position and a different action space.
Whole-body pose alongside the hands means the data carries the balance and locomotion context a humanoid action depends on, not just the manipulation in isolation.
Bimanual capture
Two wrist cameras plus 3D hand pose keep both hands in frame with per-hand pose and visibility flags throughout coordinated, often asymmetric two-handed work. Body pose and IMU add the whole-body context: stance, reach, and transfer between locomotion and manipulation.
Instance segmentation and action segments mark what each hand is doing and when, so a policy can learn the handoff between the two.
The spec
You describe the tasks your humanoid needs to perform. We turn it into a written spec covering tasks, environments, regions, capture tier, annotation depth, and accept criteria, then quote reach, timeline, and price against that document before any capture begins.
The collection runs through the same seven-stage end-to-end custom collection pipeline as every Firsthand program: contributor recruiting, capture, per-batch QA, anonymization, and licensed delivery. Rejected items are recaptured at our cost.
Fast · on demand
Where coverage is already deep, the first validated batch lands in a median of 48 hours from signed spec. Thin environments typically take three to four weeks, because vetting operators there is the bottleneck, and we quote that timeline up front.
On-demand top-ups run against the same spec and accept criteria after each training run, so new humanoid training data stays consistent with what you already hold.
Rights
Every custom collection ships under a perpetual, buyer-owned commercial license, and exclusivity is available so the data is never resold. Each episode links to a signed contributor release through a documented consent chain.
Faces, plates, and screens are detected and irreversibly blurred at ingest, and a reviewer confirms the result before anything ships. The anonymization record is delivered with the data.
Humanoid need vs delivery
| Humanoid training need | What Firsthand captures |
|---|---|
| Head camera viewpoint | 4K head-mounted camera matching a head-worn humanoid perception point |
| Two-hand manipulation | Two wrist cameras with per-hand visibility, capturing bimanual and asymmetric work |
| Depth | Synchronized depth stream for scene and contact geometry |
| Whole-body motion | IMU plus 3D body pose for stance, reach, and locomotion context |
| Hand pose | 3D hand pose per hand, aligned with the wrist camera streams |
| Action structure | Instance segmentation, action segments, grasp and contact events |
| Quality | Measured against signed accept criteria; billed per validated hour; reject log with every batch |
| Formats | RLDS, LeRobot, WebDataset, plus HDF5, zarr, and Rerun .rrd |
Every collection runs through the same seven-stage end-to-end custom collection pipeline.
FAQ
Sources
Tell us the tasks your humanoid needs to perform. We will quote reach, timeline, and price against a written spec, with a first validated batch in a median of 48 hours where coverage is deep.