Egocentric Fitting Room and Apparel Merchandising Data

Egocentric fitting room and apparel merchandising data is first-person video of an associate handling garments and retail displays, recorded from their own point of view with synchronized sensor streams alongside. The camera sits on the head, close to what a robot’s onboard camera would see, so the frames show what the hands do with the hanger, the garment and the shelf. The same idea sits behind any apparel retail robot training data or fitting room reset video dataset request, applied here as a custom collection scoped to your tasks.

The data is human capture: a paid, consented contributor wears the rig and performs a task from a written skill spec. It is not robot or teleoperation data. It is also not fixed-camera store footage, because the viewpoint is the person doing the work.

The task catalog lists six rows. Three sit in Retail Store -- Front of House, in the Retail & Consumer Goods sector, which the catalog illustrates with a clothing store, an electronics shop, a furniture showroom, sporting goods, a bookstore and a department store: "Fitting room reset", "Garment rail sizing" and "Merchandising and facing". Three sit in Retail Floor Set, illustrated with shelving, a checkout, a garment rail and a pack bench: "Reset fitting room", "Hang + size-order rail" and "Fold + stack apparel". The rows table below gives each description, skill level and status.

The "Merchandising and facing" row speaks of products, not garments. It is counted here as a store-floor display task, and the page does not claim it is specific to apparel.

No. The three Front of House rows use a head-mounted phone with a multiview rig. The three Retail Floor Set rows use a multiview rig with handheld grippers instead. The first-person view of real work comes from the head-mounted phone, and the multiview rig adds synchronized third-person angles when the spec asks for them.

The site states no garment categories, size ranges, rail lengths, store formats, store chains or facility names for these rows.

Several rows sit next to these without being part of them. Laundry folding and ironing are garment-handling tasks with their own pages. "Price-tag items" attaches price tags or labels and has its own page. Returns processing, which can involve garments, is covered by the returns page. Generic restocking rows fill shelves from cartons or a cart. This page does not count any of them as fitting room or apparel merchandising rows.

No. No catalog row names helping a customer try on or choose a size, answering fitting-room requests, working in planogram or inventory software, tracking sales, or loss prevention. The coverage table below sets each of these against what the catalog actually lists. Anything else is written into the skill spec, and feasibility is confirmed when it is quoted.

Each attempt is delivered as one episode. On this site the standard episode carries action segments, 3D hand pose for both hands with a per-joint visibility flag, contact events that mark when a hand meets an item, an objects list, one task instruction, and per-frame instance masks stored at the depth resolution of 848×480.

For apparel work, an object in the list might be a garment, a hanger, a rail or a shelf product, as written in the skill spec.

Action segments divide an episode into start- and end-timestamped spans, each carrying start_ns, end_ns, verb and noun. For apparel work, a segment might be one garment rehung, one hanger spaced or one fold completed, as written in the skill spec.

Validation runs against signed accept criteria written into the spec before capture, and a reject log ships with each batch. A human reviewer checks every episode. Only footage that clears the criteria is billed as a validated hour.

Delivery is in the format your stack already reads: RLDS, LeRobot, WebDataset, HDF5, zarr or Rerun .rrd, or raw streams with metadata. The license is perpetual, irrevocable and worldwide, buyer-owned and non-exclusive by default. The executed agreement controls.

Garments are soft, so a hand closes on them and the cloth changes shape. A hanger has to meet a rail, a stack has to stay square, and a rail has to be read for size and style before the next garment is placed. The hands are often partly hidden by the cloth they hold. That is why hand pose carries a per-joint visibility flag, why contact events mark when the hand meets the item, and why masks are recorded per frame. The challenge table below maps each difficulty to the stream or label this site states.

The catalog describes the head-mounted phone as recording the first-person view of real work. The site also states that Firsthand does not capture at a customer’s own facility.

Every contributor signs a written release, and each delivered episode links to it through consent artefact IDs. At ingest, automated detection finds faces, plates and screens and blurs them irreversibly, and speech is scrubbed of names, addresses and card numbers. A human reviewer confirms that no un-consented person is identifiable in any frame, and the anonymization record ships with the data.

The site states no rule about where a camera may point near a fitting room. Any such constraint is written into the spec. Certifications are not stated for these environments, and this page makes no such claim.

From signed spec to first validated batch, the published timeline is a median of 48 hours where coverage is deep and three to four weeks where it is thin and contributors need to be recruited first. All six rows are marked Available on request, so each starts from the spec rather than from an already prepared setup.

Describe the task and its order, the garments or products involved, where each one ends up, what counts as a finished step, and the failure or variation cases your pipeline is missing. If you need something beyond the six rows, such as helping a shopper or working in a planogram tool, say so up front. That becomes a written skill spec, quoted for reach, timeline and price before any capture begins.

Most programs begin with a Spec Pilot, a small, fast first batch that proves the spec and the accept criteria on real footage before volume scales. Collections are billed per validated hour.