Egocentric Plastic Chair Manufacturing Data

It is first-person video of a person doing the work of a plastic chair line, recorded from the contributor’s 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 while it handles a part, so the frames show the hands on the shell, on the legs and on the hardware, rather than a distant view of someone at a line. On this site it is a custom collection: you name the task, the setting and the variation, and capture is run to a written skill spec.

The data is human capture. A paid, consented contributor wears the rig and does the task. It is not robot or teleoperation data, and it is not a fixed corpus you download. People reach this topic through close cousins: furniture manufacturing robot training data, an egocentric molded chair assembly video set, a flash trimming and leg fitting dataset, or custom plastics factory data collection. They describe one request from different angles, and the work is the same: write down the task, collect it on demand, and receive validated episodes.

The site’s catalog row is about plastic chairs. It does not list other kinds of furniture under this row, and this page does not extend the row to them. Nothing on this page states that the video carries a flash label, a fit check, a chair-quality score or a measurement.

One. "Plastic Chair Manufacturing" sits in the Plastic & Light-Assembly Manufacturer environment of the Manufacturing & Workshops sector, whose listed examples are consumer goods assembly, packaging operations, labeling and quality inspection on a production line. The catalog describes it as: "Unload molded chair shells, trim flash, fit legs and hardware and stack finished chairs." It is rated Medium, lists a head-mounted phone or a multiview rig as the capture setup, and is listed as Ready to deploy.

The catalog defines Ready to deploy as contributors and site access already set up. The site says nothing further about what that status means in volume, hours or delivered footage, so this page claims no completed collection and no delivered hours, and what applies to a given order is confirmed when you request the quote. This page names only the objects and steps that appear in the row: molded chair shells, unloading, trimming flash, fitting legs and hardware, and stacking finished chairs. The first table below reproduces the row as the catalog states it.

The egocentric injection molding machine tending page is about the two rows that open the mold press and remove parts. The egocentric part trimming and deflashing page is about the two Easy rows that trim flash from molded parts with flush cutters and a deburring blade. The egocentric light assembly page covers all fifteen rows of the Plastic & Light-Assembly Manufacturer environment. The chair row appears on those pages only as one listing among others, and none of them has it as its main topic.

This page is the one place that treats the chair row as the subject: a four-step chain, from unloading to stacking, with trimming as only one of the steps. The site’s hardware and light-assembly plants page lists a separate furniture plant type whose work is fitting panels, drilling and fastening and finishing. That is a different environment from this row, and this page does not claim it. The second table below sets the pages side by side.

A chair line is a sequence in which each step hands the work to the next, and the hands are present in all of them. The site states that egocentric capture sits close to what a robot’s onboard camera would see, and that 3D hand pose carries what grip and finger changes look like when plain video cannot resolve them. Human capture does not replace robot data, and this page does not claim that it does.

A custom collection also lets you aim at the step your model is missing, such as the leg and hardware fitting, rather than hoping a broad collection happens to contain it. The site lists no off-the-shelf plastic chair manufacturing dataset, so this work is custom collection, and this page does not claim that any public corpus lacks chair-line data.

Each attempt is delivered as one episode, and a chair task needs a decision up front about what one attempt is: one chair from shell to stack, one step such as the leg fitting, or a run of chairs. The site does not set that length, so the skill spec does. Inside the episode, the synchronized streams run from start to end, and action segments divide it into labeled spans.

An episode might open with a shell being unloaded and close with the finished chair being stacked, as the catalog row words the work. That is an illustration of how a spec can cut an episode, not a stated duration. How many chairs are covered is set in the spec. The site states that Firsthand does not capture at a customer’s own facility, so the line or bench used is agreed in the skill spec.

A chair collection uses the same capture stack as every Firsthand program. The instrumented tier is a head-mounted stereo rig that records first-person RGB, with a metric depth stream recovered from the calibrated stereo rig rather than from a standalone depth sensor, and 6-DoF IMU data at 200 Hz from head- and wrist-mounted sensors. The wider stack is ten streams: a 4K head camera, two wrist cameras, depth, IMU, 3D hand and body pose, instance segmentation and action labels, hardware-aligned under 2 ms.

Every stream is hardware-triggered onto one PTP clock domain, so depth, motion and hand pose line up with the same video frame rather than being approximated afterwards. The rig is calibrated twice per session, before capture starts and after it ends, and an episode is accepted only if RMS reprojection stays under 0.28 px. The catalog lists a head-mounted phone or multiview rig for the row; a multiview rig adds synchronized third-person angles when the spec asks for them. Which tier applies to your program is set in the skill spec.

The site does not state a camera count, resolution or frame rate specific to chair manufacturing. Force and tactile readings are not among the streams, and no robot or teleoperation data is captured.

Inside each episode, action segments divide the recording into start- and end-timestamped spans, each carrying start_ns, end_ns, verb and noun. For this task a segment can be one step as written in the skill spec, so the catalog row becomes a run of labeled steps such as unload, trim, fit and stack rather than one long clip. The other standard fields are hand_joints and hand_visible for 3D hand pose, contact_events for the moment the hand meets an object, objects, and instance_masks, stored at the depth resolution of 848×480 so they line up with the depth stream.

Hand pose is delivered for both hands with a per-joint visibility flag for occluded fingers, 21 joints per hand in 3D. The site describes contact events as the moment the hand meets an item, and it does not say whether a shell, a leg or a piece of hardware counts as the item, so the spec is the place to settle what a contact event means for your task.

The site does not list labels for the flash, the fit or the finished chair. There is no stated label for flash location, leg or hardware fit, chair quality, dimension, tolerance, defect or task-completion score, and no contact force reading. The catalog description names the work the contributor performs, not a delivered measurement.

Several limits the site already states apply directly to a chair line, and several it leaves open. Each stated limit below appears elsewhere on the site rather than being introduced here.

Hand occlusion is the first. Fingers disappear behind objects in manipulation, which is why 3D hand pose carries a per-joint visibility flag, and the contact events and per-frame masks sit alongside it. Depth is the second. The site says depth degrades on wet and reflective surfaces and handles that with per-frame confidence maps. It does not say how depth behaves on a glossy molded shell in particular.

The third is the set of machine-checked reject criteria. An episode is rejected for a sync breach beyond the ceiling for more than three consecutive frames, calibration drift beyond tolerance, 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, or a mismatch against the named skill spec. A human reviewer then checks each episode against the task spec. For a chair task, the spec decides what the manipulated object is: the shell, the legs, the hardware or the finished chair.

Also not stated: a mask accuracy figure for chair parts, and any description of how a part fully covered by the hand is handled. If your policy depends on either, say so at the start.

Validation runs against signed accept criteria written into the spec before capture. Each batch goes through per-batch QA and a reject log ships with it. Only footage that clears the criteria is billed as a validated hour, and rejected items are recaptured at our cost. Many teams start with a Spec Pilot, a small, fast first batch that proves the spec and the accept criteria on real footage before volume scales.

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

On safety and quality, the site states no contributor safety protocol, no guidance on protective equipment, no machine or press safety measure, no cycle-time, throughput, tolerance or defect-rate figure, and no material-grade, safety, quality-system or certification claim for this row, and this page does not describe one. None of this is legal advice, and the executed agreement controls.

From signed spec to first validated batch, the timeline is a median of 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. Speed comes from a warm pipeline, not from cutting consent, review or anonymization. The row is marked Ready to deploy, but the site does not say whether the plastic and light-assembly environment counts as deep or thin coverage, so ask which timeline applies when you request the quote. After the first batch, on-demand top-ups run against the same accept criteria, so you can add more chairs or steps without rewriting the spec. The site does not state a turnaround for those later batches, so plan them in your quote.

This is custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs. Delivery is in the format your stack already reads: RLDS, LeRobot, WebDataset, HDF5, zarr or Rerun .rrd, or raw streams with metadata. Readable converter source ships alongside the data so the schema can be retargeted to your own loader. Collections are billed per validated hour.

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. The site does not publish a price list, a minimum order or a batch size for this task, so those are set in your quote.

Where this site does not state something, this page says "not stated" rather than filling the gap. The fourth table below separates what is delivered from what is not stated. There is no listed label for flash, fit or chair quality, no force or tactile stream, no shell, leg or hardware specification, no equipment or tool specification, and no throughput, cycle-time or certification claim. Nothing here claims a training outcome or a model improvement, and no delivered chair-line footage or hour count is claimed.

The tradeoff to plan for is the embodiment gap between a human hand handling a shell and fitting hardware and a robot platform doing the same, which retargeting or pairing with robot data has to close. Counts of contributors, hours and chairs are set in your skill spec.

Describe the four steps and where each starts and ends, the setting, and the variation or failure cases your pipeline is missing. That becomes a written skill spec, quoted for reach, timeline and price before any capture begins, and most programs run a Spec Pilot first. The fifth table below maps the parts of a skill spec onto a chair request. A published sample recording is on the sample data outputs page if you want to see the streams before you write a spec.

Three choices matter most for this task. Name each step rather than a broad activity such as "assembly", because that is what lets each step become one action segment. Name the setting and the capture tier, since the catalog lists a head-mounted phone or multiview rig for the row. And name where the episode starts and stops: one chair, one step or a run of chairs, since the site does not set an episode length.