Egocentric Poly-Bagging and Multi-Item Bundling Data

It is real-world, first-person video of the hand work at the pack end of a fulfillment line, recorded with depth and 3D hand pose and the other streams the site already lists. The catalog describes the work in short phrases: bag an item and apply a tag, kit and bundle a multi-item order, and weigh, verify and seal a package. The buyer sets the task and the accept criteria; Firsthand handles contributors, capture, QA, consent and delivery.

The data is human capture by paid, consented contributors, not robot or teleoperation data. This page makes no claim that any robot has been trained on it. People reach this topic through several phrasings, such as egocentric poly-bag packing data, first-person multi-item bundling video, a weigh and seal fulfillment dataset or custom warehouse packing data collection. They describe one request from different angles.

Three rows in the Warehouse Zone environment name this work. Poly-bag + tag is rated Easy and reads "Bag an item and apply a tag." Bundle multi-item is rated Easy and reads "Kit and bundle a multi-item order." Weigh + seal is rated Easy and reads "Weigh, verify, and seal a package." All three are listed with the capture setup "Multiview rig + handheld grippers" and the status Available on request.

Available on request means the work is scoped and recruited to spec when a buyer asks for it. This page therefore claims no completed collection and no delivered hours for any of the three rows. The first table below reproduces the rows exactly as the catalog lists them, with the other Warehouse Zone rows shown as nearby context.

The row is one line: bag an item and apply a tag. The site states no bag sizes, bag types, tag types or item list, so a spec should name the item groups, how the item goes in, what the tag is and where an episode ends. Whether an episode is one item, a short run of items or a full bench shift is a spec decision, because the site states no episode length for this row.

The footage matters most at the moments when the bag is open and the fingers are inside or behind it, and when the tag goes on. Those are the moments hand pose with visibility flags and contact events are meant to cover. Whether the tag is read, printed or only placed is not stated, and this page does not describe it.

The row reads "Kit and bundle a multi-item order." It is the only one of the three that involves several different items at once, so the object definition is the main spec question: whether the bundle, each item or the order as a whole counts as the thing being handled, and how that choice shows up in contact events and instance masks.

The site states no number of items per order, no item types and no bundling method. It does not say what holds a bundle together. Those details are agreed in the skill spec, and an order that needs a specific mix of items should list it. Kitting also appears elsewhere in the catalog, for example parts to a tray at a precision assembly bench, which is outside this page.

The row reads "Weigh, verify, and seal a package." The environment examples line lists "a pack bench and scale", which is the only equipment the site names for the weighing step. The site states no scale model, no weight ranges and no tolerance, and it does not say what verify means, so the spec should say whether it refers to a reading, a check against an order or something else.

The site lists no label for package weight, seal quality or pass and fail, and it does not state a weight value as a delivered field. A scale readout may be visible on a screen in the frame, and the privacy rule below covers screens in general. This page gives no instruction on how to weigh or seal anything and describes only what the catalog row names.

The environment lists fourteen rows, twelve rated Easy and two rated Medium, all with the same capture setup and the status Available on request. Next to the three above sit Pack carton ("Pack a carton with void-fill and tape."), Apply shipping labels ("Apply and align shipping labels."), Cross-dock re-label ("Strip and re-label for re-route."), Scan + sort parcels ("Scan parcels and sort to lanes."), Each-pick to tote ("Pick individual items to a tote."), Load conveyor ("Load items onto a moving conveyor.") and Tote induction ("Induct totes onto a conveyor.").

The rest are Stack empty totes, Depalletize to shelf, Palletize boxes and Stretch-wrap pallet; the last two are the Medium rows. These are mentioned as context only. A spec that chains rows, for example a bag, a bundle, a weigh and seal and then a shipping label, is scoped against the written spec, and this page does not claim any such chain has been captured.

The catalog states one mode for all three 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. The site does not name a head-mounted first-person view for these rows, so whether one is added is settled in the skill spec. The environment itself is listed as "Conveyor, racking, pallets, totes, a pack bench and scale".

Each episode follows the same episode schema as other Firsthand custom collections, with ten streams synced under a 2.0 ms ceiling, head and wrist RGB, depth, 3D hand pose with per-joint visibility, contact and grasp events and frame-accurate action segments. The site also states instance segmentation for warehouse collections. Calibration is measured every session and shipped with the episode, with a 0.28 px RMS reprojection gate. The language data is one task instruction per episode. Camera count, resolution and frame rate for the set as quoted are Not stated.

Not stated. The catalog lists no stereo pair for the three rows. Stereo applies only through the site general stereo offer, which describes a stereo head rig, dataset 06, with dual fisheye cameras at 1080p per eye on a 62 mm baseline, per-unit calibration, a 9-axis IMU and hand-tracking output shipped as MCAP, and a hard-environments stack. Whether a poly-bag or bundling order uses either is a choice made in the skill spec and confirmed on a Spec Pilot.

If stereo is the point of the order, the stereo order packing and taping page describes stereo for a different set of warehouse rows. This page is about the catalog rows as listed and does not describe existing stereo footage of them.

The egocentric packing and kitting page treats packing as a task family across warehouse, manufacturing, retail, food and office settings, and it lists these three rows among many. This page makes the three rows the main topic. The egocentric box and parcel handling page covers palletizing, carrying, sealing, scanning and sorting, and Weigh + seal appears there as one row.

The envelope stuffing and mailer packing page covers an office desk, with documents and mailers, at the Office Document Station. The stereo order packing and taping page is stereo and covers four rows in the Warehouse & Distribution Center, all Ready to deploy. The second table below puts the pages side by side with their status as the site lists it.

Hand occlusion is the first case. A bag held open hides the item and the fingers, and a bundle held in both hands hides the tag. The site states that the hands occlude the work and that fingers disappear behind objects. The per-joint visibility flag marks hidden joints instead of guessing them, contact events mark when a hand meets an item, a bag or a seal, and a spec can ask for wrist or multiview angles so a grip the main camera loses is still recorded.

Reflective items are the second. The site states that depth degrades on reflective surfaces, which per-frame confidence maps handle by masking low-confidence regions. How depth behaves on a transparent or glossy bag, or on a specific item, is Not stated, so a Spec Pilot checks it on real footage. Deformable bags are the third: the site states that masks for deformable objects are shipped per frame rather than propagated, with instance masks stored at 848×480, and what counts as the object, the bag or its contents, is a spec decision.

Belt motion, crowded tables and lighting are the rest. The environment lists a conveyor, and the nearby Load conveyor row mentions a moving one, but none of the three rows names belt work and the site states no belt speed. Streams share one hardware clock, so any moving item stays aligned across streams. The site says nothing about table crowding, and an undocumented bystander is a stated reject criterion. It states no lighting values for this set, so conditions are fixed in the skill spec.

Every batch is validated against signed accept criteria before it ships, with a reject log. The site states that an episode is rejected when the manipulated object is out of frame for more than 15 percent of it, when more than 35 percent of depth pixels in the manipulation region are invalid, when a sync breach runs beyond the ceiling for more than three consecutive frames, when an undocumented bystander appears, or when the episode does not match the named skill spec. Failed episodes are rejected and logged per batch, not delivered with a caveat.

Billing is per validated hour, and rejected items are recaptured at our cost. The site does not state an accuracy figure for step labels, and it lists no label for item identity, bag contents, package weight or seal quality. What counts as a pass for a step boundary, such as the end of a bagging step or the start of a seal, is set in the accept criteria.

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, and a reviewer confirms the result before anything ships, with the anonymization record delivered alongside. Footage that cannot be de-identified is discarded.

The site states no rule specific to shipping labels, tags or a scale readout, so blurring of those is Not stated beyond the general rule above, and it is raised in the spec if it matters. The buyer owns the dataset under a perpetual, irrevocable, non-exclusive-by-default licence, with exclusivity available, and the executed agreement controls. The anonymization and privacy pages are linked below.

Delivery formats are RLDS, LeRobot and WebDataset, plus HDF5, zarr and Rerun .rrd. Firsthand also offers MCAP delivery. MCAP is an open-source container format with its own official site at mcap.dev, a specification at mcap.dev/spec and a repository at github.com/foxglove/mcap; this page describes the container only from those sources. Only dataset 06 is stated as shipped as MCAP, and for any custom order MCAP can be the only delivery or sit beside a training format.

Topic names, message schemas, encoding and rates are Not stated on the site and are agreed in the skill spec. Converter source is readable, and the episode schema page lists the fields, so a loader that already reads another Firsthand dataset reads this one without changes.

Carrier, marketplace, retailer and packaging brand names, bag and tag types, item lists, items per bundle, scale models, weight ranges, pack rates, task times, throughput, success and error rates, and force or tactile data are not stated on the site. None is claimed here.

If your policy needs any of the fields that can be specified, they are written into the skill spec as annotation depth and covered in the quote before capture begins.

The site off-the-shelf catalog lists about 324,800 hours across 18 datasets, including about 64,800 hours of stereo video in 9 datasets. A few of its environment descriptions, such as S2, S5 and M7, mention packing or packaging among other factory work, but none names poly-bagging, tagging, bundling or weighing and sealing. No off-the-shelf set is described as covering these three rows, so this is custom collection.

Custom warehouse packing data collection runs as custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs. Where coverage is deep, the median time from signed spec to first validated batch is a median of 48 hours; where it is thin, turnaround runs three to four weeks. All three rows are Available on request, so the scoped-to-spec timeline applies, and the quote states it up front.

Describe the rows you want and where each starts and ends. Say whether an episode is one bagged item, one bundle, one weigh and seal, or a chain of them, which item groups and bags are in scope, what counts as the object, what counts as a reject, and whether you want the catalog multiview rig, wrist views or a stereo pair. Then name the streams you want in the file and the delivery format.

Most programs begin with a Spec Pilot, a small first batch that proves the spec and accept criteria on real footage before you scale. The last table below shows what each spec element decides.