Now recruiting capture teams worldwide. Get paid to record first-person video of real work.

Apply to capture

Hard task

Egocentric Industrial Garment Sewing Data

Feeding fabric pieces through an industrial sewing machine to join seams and attach parts is one row in the site’s task catalog, rated Hard. This page covers egocentric garment sewing data for that one row: first-person industrial sewing machine video, what a seam joining robot dataset needs from the capture, and what the site does and does not state. We run custom apparel production data collection fast and on demand against a written skill spec, as custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs.
Catalog row
Garment sewing, in Textile & Apparel Production
Skill and status
Hard; Available on request, as the catalog lists it
Catalog capture setup
Phone (head-mounted) / multiview rig
License
Perpetual, buyer-owned

Why first-person here

What does a first-person view add to industrial sewing machine video?

01

The hands, the fabric and the machine in one frame

A head-mounted view keeps both hands, the fabric pieces and the industrial sewing machine in frame, close to what a robot’s onboard camera would see at the machine. Hand pose, contact events and per-frame instance masks are recorded on one clock.

02

Fingers hidden by the cloth, flagged not guessed

The site states that fingers disappear behind objects and that the hands occlude the work. 3D hand pose carries a per-joint visibility flag, so a finger hidden under a piece of fabric is marked as hidden rather than silently filled in.

03

Cloth that changes shape, masked per frame

The site says masks for deformable objects are shipped per frame rather than propagated from one frame to the next. Fabric pieces are deformable, so the instance masks follow them frame by frame.

04

Joining and attaching as separate steps

The row names two outcomes, joining seams and attaching parts. A spec can name each as its own step, so each becomes its own labeled action segment on one timeline.

Definition

What is egocentric industrial garment sewing data?

It is first-person video of a person doing the work named in one catalog row, feeding fabric pieces through an industrial sewing machine to join seams and attach parts, recorded from their own point of view with synchronized sensor streams alongside. People reach it through several phrasings: egocentric garment sewing data, first-person industrial sewing machine video, a seam joining robot dataset, or custom apparel production data collection. They describe one request from different angles.

The data is human capture. A paid, consented contributor wears the rig and performs the task to a written skill spec. It is not robot or teleoperation data, so using it for a specific robot still needs your own retargeting step, and it is not a fixed corpus you download: none of the site’s off-the-shelf dataset descriptions names garment sewing, so this is custom collection.

Catalog row

Which catalog row does this page rely on, and what is its status?

One row, in the Textile & Apparel Production environment of the Manufacturing & Workshops sector, whose examples line reads "Garment factory, sewing workshop, fabric cutting facility, textile mill, fashion manufacturing". It is named "Garment sewing" and reads "Feed fabric pieces through an industrial sewing machine to join seams and attach parts." Its skill is Hard, its capture setup is Phone (head-mounted) / multiview rig and its status is Available on request. The first table below reproduces it.

Available on request is the status the catalog gives. This page does not call the row active or ongoing, and it claims no volume, hours or delivered clips for it. What the status means for a given order is confirmed when you request the quote. The row names a machine, a material and two outcomes: an industrial sewing machine, fabric pieces, seams and parts. It names nothing more specific, so anything finer goes into the skill spec.

Hard row

Why is the Garment sewing row rated Hard?

On this site, Hard is the skill level shown in the catalog’s Skill column, one of Easy, Medium or Hard, and this page uses the word in no other sense. The row carries it. The hard-task hub lists every row rated Hard, grouped by environment, and this row is one of them. In the same environment, the Fabric cutting and Textile mill operation rows are rated Medium and Finishing and pressing is rated Easy.

What the row text itself shows is a description of one continuous action on a machine: pieces of fabric are fed through it, which joins seams and attaches parts. The site publishes no failure rate, tolerance or claim about workers’ skill for this row, so this page makes none.

Difference

How is this different from the stereo garment sewing and fabric cutting page and the egocentric sewing page?

The stereo garment sewing and fabric cutting page covers all four Textile & Apparel Production rows as a stereo and MCAP question, so Garment sewing is one of four rows there. The egocentric sewing and textile handling page is wider still: it covers a textile arts and stitching studio, the apparel production setting, leathercraft rows and a cloth bench. The egocentric garment production page covers the same four production rows as first-person capture. None of them has the Garment sewing row as its main topic.

This page is about that one row only, in first-person framing, which is why the hard-task hub now points the row here. The comparison table below sets the pages side by side.

Capture setup

What capture setup does the row list?

The catalog lists "Phone (head-mounted) / multiview rig" for the row. The catalog describes the head-mounted phone as recording the first-person view of real work, and a multiview rig adds synchronized third-person angles when the spec asks for them. The catalog does not say which of its two modes applies to a garment sewing program, so that is confirmed in the skill spec. A third-person angle can be useful where the machine is only partly visible from the head camera.

The site also describes a ten-stream instrumented tier: a head-mounted stereo rig worn by the contributor, with first-person RGB and metric depth recovered from the calibrated rig. Whether a garment sewing program uses it is Not stated. The site lists capture at a customer’s own facility as not offered, so the setting is agreed in the spec. The capture setup table below lists the figures the site states.

Episodes

What does each episode contain?

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, and per-frame instance masks stored at the depth resolution of 848×480.

Force and tactile readings are not among the listed streams, and neither is any label for whether a seam is joined acceptably. If your model needs one of those, say so in the spec before capture, because the site does not promise it.

Why it is demanding

What makes feeding fabric through an industrial sewing machine demanding to capture?

The reasons come from the row text and existing site copy only. First, the row is rated Hard in the catalog. Second, the row describes fabric pieces moving through a machine, so cloth, hands and equipment share the frame. Third, the site states that the hands occlude the work and that fingers disappear behind objects. That is why hand pose carries a per-joint visibility flag, why contact events mark when a hand meets an item, and why masks for deformable objects are shipped per frame.

The site also states that depth degrades on wet and reflective surfaces and that metric depth on sub-centimetre parts is flagged rather than interpolated. Whether anything at an industrial sewing machine counts as such a surface or part is Not stated, and so is a mask accuracy figure.

Steps

Which steps can a spec label?

The row gives two outcomes, not a list of steps, so the steps come from the spec. A spec that says "sew the garment" gives one long action segment. A spec that names each part you want separated, such as feeding the pieces, joining a seam and attaching a part, gives a sequence you can slice by step, each with start_ns, end_ns, verb and noun.

Whether an episode covers one seam, one attached part or a longer run is a spec decision, not a default. The site sets no episode length for this row. The catalog names no sub-step, so this page names none and gives no sewing procedure. What counts as a finished seam or an attached part is Not stated and is written into the spec as accept criteria before capture.

Labels and QA

How is the data validated?

Validation runs against signed accept criteria written into the spec before capture, and a reject log ships with each batch. Episodes are machine-checked against stated reject criteria, including a sync breach, calibration drift, the manipulated object out of frame for more than 15 percent of the episode, more than 35 percent invalid depth pixels in the manipulation region, an undocumented bystander, and a mismatch against the named skill spec. A human reviewer then checks each episode.

Only footage that clears the criteria is billed as a validated hour, and rejected items are recaptured at our cost. For this row, the spec decides what the manipulated object is: the fabric pieces, the part being attached or the machine.

Not stated

What does the site not state about garments, machines and results?

Garment types, fabric types, machine models, brands and speeds, seam types, measurements, cycle times, error rates and what a correctly joined seam means are Not stated. They are set in the written spec, and this page claims none of them. The site states no factory line, throughput or production rate for this row. It also states no result from training any model on this data, and names no factory, operator, brand or customer for it. It states no contributor safety protocol, licence or certification for any sewing row, and this page describes none.

Privacy

How are people and items in frame handled?

Every contributor signs a written release, and each delivered episode links to it through consent artefact IDs in the batch data card. The consent chain has six checkpoints, and a break at any one discards the footage. At ingest, automated detection finds faces, plates and screens and blurs them irreversibly. Speech is scrubbed of names, addresses and card numbers, while ambient audio is kept. A reviewer confirms that no un-consented person is identifiable in any frame, and footage that cannot be de-identified is discarded. The anonymization record ships with the data.

Who may be in frame, and how garments or patterns belonging to a facility are handled, is agreed in the spec. This is not legal advice, and the executed agreement controls.

Scale and speed

How do custom, fast and on-demand collection apply to garment sewing?

The October 2026 off-the-shelf catalog lists ~324,800 hours across 18 datasets. None of the site’s off-the-shelf dataset descriptions names garment sewing, so this is custom collection, not an off-the-shelf purchase.

It is custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs. From signed spec to first validated batch, the 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. Which applies to this row is confirmed when the spec is quoted rather than assumed, and because the row is Available on request, ask about it when you request the quote. Collections are billed per validated hour, and the site publishes no price list, minimum order or batch size for this row, so those are set in your quote.

Delivery

Which formats and license come with the data?

Delivery is in the format your stack already reads: RLDS, LeRobot, WebDataset, HDF5, zarr or Rerun .rrd, or raw streams with metadata, with readable converter source alongside so the schema can be retargeted to your own loader.

The license is perpetual, irrevocable and worldwide, with the data buyer-owned and non-exclusive by default. Training, evaluating, fine-tuning and deploying models, including commercial ones, is permitted, and exclusivity is available as a separate term. The executed agreement controls.

Ordering

How do you order a custom garment sewing dataset?

Describe where each attempt starts and ends, what counts as a finished step, and whether an episode covers one seam, one attached part or a longer run. Add the setting and lighting, the privacy rule for the workshop, and the variation your pipeline is missing. Then name the streams you want and the delivery format.

Most programs begin with a Spec Pilot, a small, fast 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. For the full list of rows rated Hard, see the hard-task hub linked below.

At a glance

What you get.

Catalog row
"Garment sewing": "Feed fabric pieces through an industrial sewing machine to join seams and attach parts." Textile & Apparel Production, Manufacturing & Workshops; Hard; Available on request
Catalog capture setup
Phone (head-mounted) / multiview rig
Episode contents
Action segments, 3D hand pose for both hands with a per-joint visibility flag, contact events, objects list, per-frame instance masks stored at 848×480
Sync and calibration
Hardware-triggered onto one PTP clock domain, aligned under 2 ms; calibration twice per session under a 0.28 px RMS gate
Delivery
RLDS, LeRobot, WebDataset, HDF5, zarr, Rerun .rrd or raw streams with metadata
Garment types, fabrics, machine models, seam labels, cycle times
Not stated; agreed in the skill spec
Setting
Agreed in the skill spec; capture at a customer’s own facility is not offered
License
Perpetual, irrevocable, buyer-owned, non-exclusive by default

Every collection runs through the same seven-stage end-to-end custom collection pipeline. End-to-end custom data collection is a managed service that takes an AI data need from problem to owned dataset in a single accountable pipeline.

Catalog row

Which catalog row covers industrial garment sewing?

The row relied on by this page, with the catalog description, environment, skill, capture setup and status as listed.
Task (as listed)Description (as listed)EnvironmentSkillCapture setupStatus
Garment sewingFeed fabric pieces through an industrial sewing machine to join seams and attach parts.Textile & Apparel ProductionHardPhone (head-mounted) / multiview rigAvailable on request

Compared pages

How does this page differ from the neighbouring sewing and garment pages?

Each page, its main topic and how it treats the Garment sewing row. “Not covered” means the page leaves it out.
PageMain topicCapture framingTreatment of the Garment sewing row
This page: egocentric industrial garment sewing dataThe Garment sewing row onlyFirst-person, head-mounted phone or multiview rig as the catalog lists itThe whole page: scope, limits, labels, steps and ordering
Stereo garment sewing and fabric cutting dataThe four Textile & Apparel Production rowsStereo pair, agreed in the skill spec; not stated on the rowsOne of four rows, treated as a stereo and MCAP question
Egocentric garment production video dataThe same four Textile & Apparel Production rowsFirst-personOne of four rows in a task-family overview
Egocentric sewing and textile handling video dataCutting, fitting, hemming, stitching and fastening fabric across several environmentsFirst-personOne row among many, listed in a production-rows table
Hard task data collectionEvery catalog row rated Hard, grouped by environmentAs each row lists itRow listed in the hub table with its status and capture setup

Capture setup

What is the capture setup for a garment sewing collection?

The capture setup as stated on this site. Counts of sites, contributors, episodes and hours are set in the skill spec.
PartAs stated on this siteFor this row
Catalog capture modeHead-mounted phone for the first-person view of real work; a multiview rig adds synchronized third-person angles when the spec asks for themStated in the catalog; which mode applies is confirmed in the skill spec
Instrumented tierHead-mounted stereo rig worn by the contributor; first-person RGB with metric depth recovered from the calibrated rigWhether a program uses it: Not stated
Streams (instrumented tier)Ten: 4K head camera, two wrist cameras, depth, IMU, 3D hand and body pose, instance segmentation, action labelsStated; set in the spec
SyncHardware-triggered onto one PTP clock domain; aligned under 2 msStated
Hand pose21 joints per hand, 3D, in metres in the head-camera frame; per-joint visibility flagStated
CalibrationTwice per session; accepted only under the 0.28 px RMS reprojection gateStated
MasksPer-frame instance masks, stored at 848×480; masks for deformable objects are shipped per frameStated
SettingCapture at a customer’s own facility is not offeredSetting agreed in the skill spec

Demands and streams

Which stated challenge maps to which stream or label?

Each challenge mapped to a stream or label this site already states. Where nothing is stated, the row says so.
ChallengeWhat the site statesStated stream or label
Fingers hidden by fabricThe hands occlude the work; fingers disappear behind objects3D hand pose with per-joint visibility flags; per-frame instance masks; reject if the object is out of frame for more than 15% of an episode
Fabric that changes shapeMasks for deformable objects are shipped per frame rather than propagatedPer-frame instance masks
Hand meets fabric or machineHand and work are in close contactContact events mark when a hand meets an item; no finer contact label is stated
Machine only partly visibleA multiview rig adds synchronized third-person angles when the spec asks for themCapture mode agreed in the spec; wrist cameras add a closer view of the hands
Seam or attachment qualityNo such label is listedNot stated: completion is defined in the spec and episode boundaries
Force and contact pressureNo force or tactile stream is listedNot stated

Delivered vs not stated

What is delivered and what is not stated?

Delivered items are stated elsewhere on this site. Items marked not stated are not promised on this page.
ItemStatusDetail
Action segmentsDeliveredstart_ns, end_ns, verb, noun, one per step
3D hand poseDelivered21 joints per hand; per-joint visibility flag for occluded fingers
Contact eventsDeliveredThe moment a hand meets an item
Per-frame instance masksDeliveredStored at 848×480
Consent and anonymization recordsDeliveredConsent artefact IDs per episode; anonymization record shipped with the data
Garment types, fabrics, machine modelsNot statedAgreed in the skill spec
Measurements, cycle times and throughputNot statedNo such figure is listed
Number of sites, episodes and hoursNot statedSet per program in the skill spec
Seam or attachment success labelsNot statedNo such label is listed

Delivery and licensing

What is delivered, in which formats, and under what license?

Delivery and licensing terms as stated on this site. “Set per program” means the site states no fixed value.
ItemAs stated on this siteNote
FormatsRLDS, LeRobot, WebDataset, HDF5, zarr, Rerun .rrd, or raw streams with metadataChosen in the skill spec; converter source ships with the data
LicensePerpetual, irrevocable, worldwide, buyer-owned, non-exclusive by defaultExclusivity is available as a separate term; the executed agreement controls
BillingPer validated hourRejected items are recaptured at our cost
PrivacyFaces, plates and screens blurred irreversibly; footage that cannot be de-identified is discardedAnonymization record delivered with the data
Timelinea median of 48 hours to first validated batch where coverage is deep; three to four weeks where thinWhich applies to this row is Not stated; quoted against your spec
Price, minimum order, batch sizeNot published for this rowSet in your quote

Skill spec

What does the skill spec fix for a garment sewing order?

Spec element and where a buyer can see it in the delivery. “Set per program” means the site states no fixed value.
Spec elementWhat it decidesWhere it shows up
Outcomes in scopeWhether joining seams, attaching parts, or both are in scopeWhat the cameras see; set per program, no per-outcome label is stated
Steps and orderWhat one step is and how steps are namedEpisode boundaries and action segments; set per program
Episode boundaryOne seam, one attached part or a longer runSet per program; the site does not state an episode length for this row
Object definitionWhether the fabric pieces, the part or the machine counts as the objectContact events and the out-of-frame reject rule; set per program
Capture modeHead-mounted phone, multiview angles, wrist views and any othersThe streams you can read; confirm on the Spec Pilot
Setting and privacyWhere capture happens and what may be in frameSettled before capture; bystander handling is a stated reject criterion
Accept criteriaWhat passes QA, including what counts as a rejectReject log per batch
Delivery formatHow the data is packagedRLDS, LeRobot, WebDataset, HDF5, zarr, Rerun .rrd or raw streams
LicenseWho owns the datasetPerpetual, irrevocable, buyer-owned, non-exclusive by default

Example footage

What does first-person capture look like?

One real clip already published on this site, shown as a format and rig example: ten seconds of first-person home cooking with faces blurred. It does not show garment sewing or any textile work.

egocentric_cooking_10s.mp4 · first-person view · faces blurred · 1080 × 1080 web preview · format example, not a garment sewing clip

No garment sewing clip is published on the site, so this is the nearest first-person clip, labelled as a format example. Request a sample to see delivered files. See sample data outputs, request an egocentric data sample.

FAQ

Egocentric garment sewing data, answered.

What is egocentric industrial garment sewing data?
Custom first-person video of a paid, consented contributor feeding fabric pieces through an industrial sewing machine to join seams and attach parts, captured to a written skill spec.
Which catalog row covers it, and what is its status?
One row, "Garment sewing", in Textile & Apparel Production: "Feed fabric pieces through an industrial sewing machine to join seams and attach parts." It is rated Hard, lists a head-mounted phone or multiview rig, and is Available on request. This page claims no volume or hours.
What does Hard mean here?
Only the skill level the catalog lists in its Skill column, one of Easy, Medium or Hard. The site states no failure rate, tolerance or worker skill claim for this row.
How is this different from the stereo garment sewing and fabric cutting page?
The stereo page covers all four Textile & Apparel Production rows as a stereo and MCAP question. This page is about the Garment sewing row only, as first-person video.
Is there an off-the-shelf garment sewing dataset?
No. None of the site’s off-the-shelf dataset descriptions names garment sewing. It is custom collection, at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs.
Does the data include seam quality or stitch labels?
No such label is listed. The episode carries action segments, hand pose, contact events, an objects list and per-frame instance masks. What counts as a joined seam or an attached part is defined in the spec as accept criteria.
How fast can a custom garment sewing dataset be collected?
From signed spec to first validated batch, a median of 48 hours where coverage is deep and three to four weeks where it is thin. Ask which applies when you request the quote.
Who owns the data and in which formats does it arrive?
The buyer owns the delivered data under a perpetual, non-exclusive commercial license by default. Formats are RLDS, LeRobot, WebDataset, HDF5, zarr or Rerun .rrd, or raw streams with metadata.

Sources

Where are the delivery formats documented?

Request a custom garment sewing dataset.

Tell us where each attempt starts and ends, whether an episode covers one seam, one attached part or a longer run, the setting, the privacy rule for the workshop, and the streams and format you want. We will quote reach, timeline and price against a written spec.