Egocentric Security Checkpoint Task Video Data

Egocentric security checkpoint task data is real-world video of the physical hand work done around a checkpoint: putting items into bins, moving bags onto a belt, opening and re-packing a bag, taking an item back out and stacking the bins afterwards. It is recorded against a written spec, with the streams the site states for its custom collections. The tasks are the ones the site states under its Mock Security Checkpoint catalog environment.

It is task data, not footage of people in transit. The site describes the set as staged, and this page does not describe capturing real travellers, passengers or faces. Where a row involves a person-shaped target, the catalog names a mannequin.

The site states checkpoint work in one catalog environment, Mock Security Checkpoint, under the Public Spaces & Outdoor sector. Twelve task rows sit under it. Nine are rated Easy and three are rated Medium. Every row lists the same capture mode, a multiview rig with handheld grippers, and every row carries the status Available on request.

The examples line lists the props of the set and states that no live site is involved. That is the site's own wording for a staged set, and this page repeats it rather than adding to it. Prop counts, layouts and dimensions are Not stated.

Three Easy rows cover bins. Load screening trays places bags and loose items into bins. Retrieve flagged item retrieves a flagged item from a bin. Bin stacking + return stacks empty bins and returns them. Together they form the loop of filling, emptying and resetting.

The site states no bin sizes, item lists or fill rules, and it does not say what makes an item flagged, so none of that is claimed. A spec should name the item groups, how many go in a bin and which item is retrieved.

Four rows cover bags and packages. Baggage load to belt loads bags to a belt at pitch and is rated Medium. Bag search + re-pack opens, inspects and re-packs a bag, also Medium. Package inspection opens, inspects and reseals a package, again Medium. Divest + re-collect divests belongings and re-collects them afterwards, and it is rated Easy.

The phrase "at pitch" is catalog wording, and the site states no belt speed, spacing or bag counts. Bag contents, bag sizes and what inspecting means are set in the spec. Box handling in a logistics setting has its own page, linked below.

Two Easy rows use a mannequin as the target. Handheld wand pass passes a wand over a mannequin, and Pat-down sequence (mannequin) performs a pat-down pattern on a mannequin. Using a mannequin keeps these rows free of any real person.

The site states nothing about what a wand detects, how a pass is judged or what any pattern is for. This page makes no screening-technology or threat-detection claim, and the number of passes and the pattern itself are Not stated.

Three Easy rows handle small objects. Access badge + gate handles a badge and gate object. Baggage tag scan scans a bag tag and routes it. Document check places a document under a reader. The catalog gives one line for each and no further detail.

The site states no reader, tag or badge types, no document formats and no routing rules. Tag scanning in a warehouse setting has its own page, linked below. Object types here are catalog wording only.

The catalog places the work on a staged set, and the environment page states that contributors are recruited to spec in Western program markets, so the footage reflects work done by people in North American and European businesses. The site names no venue, operator or region for this set. Access and schedule are agreed in the spec, and the number of sites is Not stated.

The privacy steps the site states are these. Every episode links to a signed contributor release through a documented consent chain, and the consent record is part of the delivery. Faces, plates and screens are detected and irreversibly blurred at ingest, a reviewer confirms the result before anything ships, and the anonymization record is delivered with the data. The site states no checkpoint-specific bystander rule, so how any other person in frame is handled is settled in the spec.

The catalog names this mode for all twelve rows: a staged working set is recorded by a multiview rig while the operator performs the task with handheld grippers, so demonstrations map onto a robot end effector. On the rig tier, every stream is hardware-triggered onto one PTP clock domain, and the rig is calibrated twice per session, before capture starts and after it ends. An episode is accepted only if RMS reprojection stays under 0.28 px.

A head-mounted first-person view is not listed for these rows, so whether one is added is settled in the spec. Camera count, resolution and frame rate for this set are Not stated, so session length and volume are quoted against your spec.

The Firsthand episode schema, stated for custom collections, has ten streams synced under 2 ms, with RGB, depth, 3D hand pose with per-joint visibility, contact and grasp events, and frame-accurate action segments. For checkpoint work, contact events mark touches on a bin edge, a bag handle or a tray, and action segments mark each step, such as load, open, re-pack, retrieve and return.

The site does not say which streams apply to the gripper mode row by row, so the stream list is confirmed in the spec. Object labels use the terms your spec sets. Force or tactile data, and success or quality labels, are Not stated for these rows. Tactile gloves are a separate capture tier with their own page.

Delivery formats are RLDS, LeRobot and WebDataset, plus HDF5, zarr and Rerun .rrd. The format table below sets each against what its official documentation states. Action segments are the step labels, and any further object labels follow the spec.

Every batch is validated against signed accept criteria before it ships, with a reject log, and delivery is rolling, per validated batch. Billing is per validated hour, and rejected items are recaptured at our cost.

The site states no robot results for checkpoint work, so this section describes what the work asks of a dataset, not what any policy achieves. The catalog describes loops. Items go into a bin, one is retrieved, the bins are stacked and returned, and bags are opened, inspected and re-packed. The same objects are handled in a different order each time through.

Bags and packages are containers that open, hold other items and close again, and loose items shift as they are placed. Retrieval means finding one item among others, and a belt adds timing to placement. The site states no cycle times, repetition counts or error rates, so a spec names the tasks, item groups and number of sequences.

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 up front. The site does not state which of the two applies to checkpoint work.

All twelve rows are listed as Available on request, so the timeline is quoted against your spec before any capture begins.

You describe the checkpoint tasks, item groups and sequence your policy needs. We turn that into a written spec covering tasks, environments, capture tier, annotation depth and accept criteria, then quote reach, timeline and price before any capture begins.

A Spec Pilot, a small and fast first batch, can prove the spec and accept criteria before you scale. Billing is per validated hour, a reject log ships with every batch, rejected items are recaptured at our cost, and every collection ships under a perpetual, buyer-owned commercial license.