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

Apply to capture

Hard task

Stereo Chef Knife Work Data

Break down produce and proteins with precise knife cuts and cook dishes on a professional line: that is the catalog’s description of one Hard culinary row, Culinary Chef Work / Knife Work, listed under Restaurant & Commercial Kitchen. This page covers that single row on the stereo and container side: what the catalog states, what a calibrated stereo pair would add as a separate choice, how MCAP fits and what the site does not state. We run custom kitchen data collection fast and on demand against a written skill spec, as stereo knife skills video and a chef knife cuts robot dataset, at scale through a network of thousands of capture providers.
Catalog row
Culinary Chef Work / Knife Work, Restaurant & Commercial Kitchen
Skill and status
Hard; Active collection, as the catalog lists it
Catalog capture
Phone (head-mounted) / multiview rig
Container
MCAP (.mcap), open source; stereo agreed in the skill spec

Requirements

What does a chef knife capture need that a general kitchen clip does not?

01

One catalog row, stated exactly

Culinary Chef Work / Knife Work: "Break down produce and proteins with precise knife cuts and cook dishes on a professional line." The row is rated Hard, lists a head-mounted phone or multiview rig and is Active collection.

02

Two hands, a blade and a changing ingredient

Knife work is bimanual: one hand guides the ingredient while the other cuts. Hand pose, contact events and per-frame instance masks are recorded on one clock, with a visibility flag on each joint for fingers that an ingredient or handle hides.

03

Stereo and MCAP as separate, spec-level choices

A calibrated stereo pair and an MCAP container are agreed in the skill spec. The row states neither, and this page does not claim them.

04

Limits stated plainly

No knife type, cut, ingredient, dish, cut size or knife angle is stated for the row. Those are written into the spec, and this page gives no instructions and no knife-safety guidance.

Definition

What is stereo chef knife work data?

Stereo chef knife work data is first-person video of a person doing the work named in one catalog row: breaking down produce and proteins with precise knife cuts and cooking dishes on a professional line. It is recorded from the contributor’s own point of view with two calibrated cameras, and with synchronized sensor streams alongside, so the frames show what the hands are doing to the ingredient with the knife rather than a distant view of someone standing in a kitchen. Searchers also phrase it as stereo knife skills video, chef knife cuts robot data, a culinary knife work dataset or custom kitchen data collection at scale. The data is the same, collected as a custom collection scoped to your task.

On this site the data is human capture: a paid, consented contributor performs the task from a written skill spec. It is not robot or teleoperation data, and the page makes no claim that any robot has been trained on it or that any robot cuts or cooks. Stereo describes the capture and MCAP describes the container. They are separate choices, and neither is stated on the row itself.

Catalog row

Which catalog row covers chef knife work, and what does it say?

One row counts. "Culinary Chef Work / Knife Work" sits in the Restaurant & Commercial Kitchen environment, in the Food & Beverage sector, whose example line reads "Sit-down restaurant kitchen, café, coffee shop, artisan bakery, patisserie". The row reads: "Break down produce and proteins with precise knife cuts and cook dishes on a professional line." It is rated Hard, its capture is Phone (head-mounted) / multiview rig and its status is Active collection. The example line is shared by every row in that environment, so it is quoted here as context rather than assigned to this task.

Active collection is exactly what the catalog states. The site says nothing further about what that means in volume, hours or delivered footage, so this page claims none, and what applies to a given order is confirmed when you request the quote. The first table below reproduces the row, with the neighbouring vegetable chopping row as context only.

What the row does not say

What does the chef knife row leave unstated?

The row says what is done, not what it is done with or to. It names produce, proteins, knife cuts, dishes and a professional line, and nothing about the kind of knife, the kind of cut, the ingredient or the dish. It gives no cut size, no knife angle and no tolerance, and it names no procedure or recipe. Whether a program records one cut, a chain of cuts or a cooked dish from start to finish is not stated either.

A buyer who needs a particular ingredient, cut, tool, start state or finished dish as an accept criterion writes it into the skill spec, and a Spec Pilot proves it before you scale. This page does not fill any of those gaps and gives no instructions for any step.

Neighbouring rows

Which nearby kitchen rows share the environment?

The catalog lists thirteen rows for Restaurant & Commercial Kitchen. One is Hard, four are Medium and eight are Easy; nine are Active collection and four are Ready to deploy. Every one lists the same capture setup, Phone (head-mounted) / multiview rig. The chef knife row is the only Hard row in the environment.

The nearest neighbour is the Medium row for chopping and portioning vegetables with a chef’s knife and a cutting board, described as washing, trimming and chopping vegetables into even pieces and portioning them into containers. It is context only on this page. Another Medium row, Cooking / Line Cooking, covers front and back of house operations including prep, plating and service, and this page does not rely on it. The second table below sets the counts out.

Why it is hard

Why is chef knife work listed as Hard?

The site gives one definition of Hard: the skill level shown in the catalog, one of Easy, Medium or Hard. It states no criteria beyond that, and this page imports no outside rubric and states no failure rate, tolerance, cut-size standard or claim about how skilled any worker is. What can be said comes from the row text and from copy elsewhere on the site.

The row text asks for two things in one task: precise knife cuts on produce and proteins, and cooking dishes on a professional line. Elsewhere the site describes knife work as exercising fine bimanual dexterity, tool grip and control, and describes the ingredient as changing shape with every cut. One hand guides while the other cuts, so the data has to keep both hands, the blade and the food in view together, and it has to follow an object whose shape is different after each cut. The challenge table below maps each difficulty to the stream or label the site already states, and says Not stated where nothing is.

Capture

Is the Culinary Chef Work / Knife Work row captured in stereo?

The catalog lists one capture mode for the row: Phone (head-mounted) / multiview rig. The row does not state stereo capture, and this page does not describe it as stereo. Stereo reaches this row only through the site’s general stereo offer: a capture choice agreed in the skill spec and confirmed on a Spec Pilot. The site states that a multiview rig adds synchronized third-person angles when the spec asks for them, and that two wrist cameras give a closer view of the hands than the head camera alone.

The site describes two stereo setups and does not say which one a knife work order would use. The off-the-shelf stereo head rig, dataset 06, is dual fisheye cameras at 1080p per eye with a 62 mm baseline, per-unit calibration, a 9-axis IMU and hand-tracking output, shipped as MCAP. The hard-environments stack is a global-shutter 4K/60 head camera, two 1080p/60 wrist cameras, a depth sensor at 848×480 and 30 fps and 200 Hz 6-DoF IMUs. Which applies to a professional line is Not stated. Capture at a customer’s own facility is listed as not offered, and which kitchens a program uses is agreed in the skill spec.

What stereo adds

What would a stereo pair add to knife work?

The site states that a stereo pair at human-eye spacing makes depth recoverable from every frame pair, and that a calibrated pair measures disparity between two views at a fixed, known baseline rather than guessing depth from a single frame. For knife work, that puts the blade, the board, the ingredient and the hands in the same capture. The site gives no depth accuracy figure for these tasks and no label for an even piece, a cut line or a finished dish, so what a buyer does with that depth is their pipeline’s question.

The site also states that depth degrades on wet and reflective surfaces, which is where a rinsed board or a steel counter sits, so kitchen episodes ship per-frame depth confidence maps that let you mask low-confidence regions. How a stereo pair behaves on a blade or a steaming pan is Not stated. Calibration is re-solved at the start and end of every session under a 0.28 px RMS reprojection gate, and streams are hardware-triggered onto one clock with a 2.0 ms cross-stream ceiling. Heat is not sensed, and force and tactile readings are not among the streams.

Episodes

What does each chef knife work episode contain?

Each attempt is delivered as one episode, and this row needs a decision up front about what one attempt is: a single cut, a chain of cuts, or a whole step on a line. The site sets no episode length, so the spec does. 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. Hand pose is 21 joints per hand, 3D, in metres in the head-camera frame.

Action segments divide an episode into spans, each carrying start_ns, end_ns, verb and noun, from a taxonomy of 97 verbs and 512 nouns, annotated to the frame and human-reviewed. Food-prep captures ship per-frame instance masks rather than masks propagated between frames, so the thing being tracked can change shape between cuts. The site lists no label for the quality of a cut: no thickness, uniformity or evenness label, no food-state label such as doneness or temperature, and no cut taxonomy beyond the verb and noun on each segment. Ask for any such label in the spec, as a scoping conversation rather than a delivered field.

Neighbouring pages

How does this page differ from the egocentric, line cooking and cutting pages?

Several pages touch the same work, and only one other takes this row as its main topic. The egocentric chef knife page covers the same single row from the first-person side: the catalog row, labels, privacy and delivery. This page covers it on the stereo and MCAP side, and neither page claims stereo is part of the row itself.

The stereo commercial kitchen line cooking page covers this row together with the vegetable chopping row and the line cooking context. It does not make chef knife work its sole topic, so this page narrows to the single Hard row. The stereo cutting and food prep page covers a different catalog row, Peel + chop vegetables, rated Medium and Available on request in the Working Kitchen Set. The comparison table below sets each page against this one, and the egocentric stereo MCAP page is the format page for the container in general.

MCAP

What is MCAP, and what can a chef knife work delivery put in it?

MCAP is an open source container file format for multimodal log data, supporting multiple channels of timestamped pre-serialized data. It is serialization-agnostic, so messages can be Protobuf, DDS (CDR), ROS, JSON and more. Its design is row-oriented and append-only, it stores schemas alongside the data, and it offers an optional index plus optional LZ4 or Zstandard chunk compression and optional CRC checksums. Those facts come from mcap.dev and the foxglove/mcap repository, both linked below. Firsthand is not affiliated with Foxglove or the MCAP maintainers.

Firsthand offers MCAP delivery, and dataset 06 is stated as shipping as MCAP. For this row the site states a stereo pair, depth with confidence maps, hand pose, IMU, contact events, action segments and calibration elsewhere. It does not state which of those are channels in an MCAP file, how they are named, what encoding they use or at what rate. No topic name, message schema or rate is stated on this page, and none should be assumed. Name the streams that matter in the skill spec and confirm them on a Spec Pilot batch. The stream map below sets each stream beside what the site states about it.

Delivery and license

Which other formats and which license apply?

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, and MCAP can ship alone or beside any of them. The episodes follow the same episode schema as every other Firsthand collection, so a loader that reads another Firsthand dataset reads this one without changes. Which container suits a lab depends on its reader and its training loop, and the site does not rank them.

The license is perpetual, irrevocable and worldwide, buyer-owned and non-exclusive by default, with exclusivity available as a separate term. The executed agreement controls. Official documentation for MCAP, RLDS, LeRobot and WebDataset is linked in the sources below.

Validation

How is the data validated and accepted?

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, such as a cross-stream sync error above a 2.0 ms ceiling, the manipulated object out of frame for more than 15 percent of the episode, or more than 35 percent of depth pixels in the manipulation region invalid, and a human reviewer then checks each one. Only footage that clears the criteria is billed as a validated hour. These are reject thresholds, not a stated pass rate.

For this row, the spec has to say which item counts as the manipulated object at each step and where the manipulation region is, because the ingredient, the knife and the cooked dish are different items at different moments. What counts as a finished step is also written down, since the site states no outcome definition. Rejected items are recaptured at our cost.

Capture and privacy

How are contributors and their work protected?

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 while ambient audio is kept. A human reviewer confirms that no un-consented person is identifiable in any frame, footage with bystanders who cannot be de-identified is discarded, and the anonymization record ships with the data. Who may be in frame, and how a kitchen’s own staff, menus or labels are handled, is agreed in the spec.

A contributor knife-handling protocol is not stated on this site, so this page describes none and gives no knife-safety or food-safety instruction. The site’s stated safety point is about the robot: human capture avoids running an early-stage robot near knives. None of this is legal advice, and the executed agreement controls.

Speed and scale

How fast and at what scale can a custom chef knife work dataset be collected?

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. The row is Active collection, but the site does not say which range applies to a knife work order, so it is confirmed when the spec is quoted rather than assumed. After a first batch, on-demand top-ups run against the same accept criteria, so you can add variation without rewriting the spec.

Chef knife work is custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs. The site states no provider count, hours or throughput for this row, and none of its off-the-shelf dataset descriptions names it, so treat it as a custom collection. Only dataset 06, the stereo head rig, is stated as shipping as MCAP, and a lab can use that stock footage to check the rig and the format before commissioning the task.

Hard task hub

How does this page fit the hard task hub?

The hard task hub reads the catalog and lists every row with a skill level of Hard: 84 rows across 28 environments, each with its stated status and capture setup. Where a task page exists for a row, the hub links to it, and this page is the link for Culinary Chef Work / Knife Work.

The egocentric chef knife page remains linked from the hub’s related links, so both sides of the row stay one click apart. A page is linked from the hub only when it states the exact row, so the hub never points at a page that is about something else.

Getting started

How do you start a pilot?

Describe what one step is: the produce, proteins and tools and how they begin, what counts as a finished step, and the variation your pipeline is missing. Then name the streams you want, whether stereo is part of the capture, and whether MCAP is the only delivery or sits beside RLDS, LeRobot or WebDataset. 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, and rejected episodes and recaptures are never billed. Counts of sites, episodes and hours are set in the skill spec. Our skill spec guide walks through how to write one.

At a glance

What you get.

Catalog row: Culinary Chef Work / Knife Work
"Break down produce and proteins with precise knife cuts and cook dishes on a professional line."; Hard; Restaurant & Commercial Kitchen; Active collection
Sector and environment examples
Food & Beverage; "Sit-down restaurant kitchen, café, coffee shop, artisan bakery, patisserie"
Catalog capture setup
Phone (head-mounted) / multiview rig, the same for every row in the environment
Stereo on this row
Not stated. Reaches the row only through the general stereo offer, agreed in the skill spec and confirmed on a Spec Pilot
Stereo head rig (dataset 06)
Dual fisheye, 1080p per eye, 62 mm baseline, per-unit calibration, 9-axis IMU, hand-tracking output; shipped as MCAP
Streams and sync
Ten streams, hardware-triggered onto one clock; 2.0 ms cross-stream ceiling
Hand pose
21 joints per hand with a per-joint visibility flag
Knife type, cut, ingredient, dish, cut size and knife angle
Not stated; agreed in the skill spec
MCAP topics, schemas and rates
Not stated; agreed in the skill spec
Delivered hours
None claimed on this page
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.

Live catalog rows

Which catalog row covers chef knife work, and which row sits beside it?

The knife work row as the catalog states it, with the neighbouring vegetable row as context only, reproduced without change. No delivered hours are claimed.
Task (as listed)Description (as listed)EnvironmentSkillCapture setupStatusRole on this page
Culinary Chef Work / Knife WorkBreak down produce and proteins with precise knife cuts and cook dishes on a professional line.Restaurant & Commercial KitchenHardPhone (head-mounted) / multiview rigActive collectionThe row this page covers
Chop and portion vegetables using a chef's knife and cutting boardWash, trim and chop vegetables into even pieces and portion them into containers.Restaurant & Commercial KitchenMediumPhone (head-mounted) / multiview rigActive collectionNearby context only

Environment mix

Where does the Hard row sit among the environment’s rows?

All thirteen catalog rows for Restaurant & Commercial Kitchen, counted by skill and status from the catalog. Every row lists the same capture setup.
SkillRowsActive collectionReady to deployCapture setup
Hard110Phone (head-mounted) / multiview rig
Medium431Phone (head-mounted) / multiview rig
Easy853Phone (head-mounted) / multiview rig
All rows1394Phone (head-mounted) / multiview rig

Comparison

How does this page compare with the egocentric sibling and the closest stereo kitchen pages?

Each page described only as the site states it. No delivered hours are claimed for any of them.
PageMain subjectKnife work row on that pageHow this page differs
This pageThe Culinary Chef Work / Knife Work row, with stereo and MCAP as spec-level choicesThe main topicStereo and container side of the single Hard row
Egocentric chef knife work dataThe same single row, first-person sideThe main topicCovers the row from the egocentric side; this page covers stereo and MCAP
Stereo commercial kitchen line cooking dataThe knife work row and the vegetable chopping row, with line cooking contextOne of two rowsWider stereo kitchen page; this page takes the Hard row on its own
Stereo cutting and food prep dataThe Peel + chop vegetables row, Medium, Available on request, in the Working Kitchen SetNot one of its rowsA different kitchen environment and row
Egocentric cutting and food prep dataFour cutting rows and the documented knife-work categoryOne of several rowsWider first-person cutting family
Egocentric stereo data in MCAPThe MCAP delivery format for stereo capture in generalNot covered; format pageApplies the format to one row
Hard-task data collectionThe hub of rows the catalog rates HardThe row is in the Hard-row table and links hereOne row out of the hub

Capture setup

What is the capture setup for the chef knife row?

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 site
Capture mode (Culinary Chef Work / Knife Work)Phone (head-mounted) / multiview rig, as the catalog lists it
Multiview rigAdds synchronized third-person angles when the spec asks for them
Stereo head rig (dataset 06)Dual fisheye, 1080p per eye, 62 mm baseline, per-unit calibration, 9-axis IMU, hand-tracking output; shipped as MCAP
Hard-environments stack4K/60 global-shutter head camera, 120° lens; two 1080p/60 wrist cameras; depth 848×480 at 30 fps; 200 Hz 6-DoF IMU
SyncHardware-triggered onto one clock; 2.0 ms cross-stream ceiling
CalibrationRe-solved at the start and end of every session; 0.28 px RMS gate
Hand pose21 joints per hand, 3D, in metres in the head-camera frame; per-joint visibility flag
Depth on wet and reflective surfacesDegrades; per-frame confidence maps ship for kitchens
Stereo on this rowNot stated; a separate offer, agreed in the skill spec and confirmed on a Spec Pilot
Which rig applies to a professional lineNot stated; confirmed in the skill spec

Capture challenges

Knife work challenge × what goes wrong × stated stream or label

Each capture challenge mapped to a stream or label this site already states. Where nothing is stated, the row says so.
Knife work challengeWhat goes wrongStated stream or label
One hand guides, one hand cutsThe hands, the blade and the food overlap and hide one another3D hand pose for both hands with per-joint visibility flags; per-frame instance masks; a wrist view alongside the head camera; reject if the object is out of frame for more than 15% of an episode
Hand meets itemThe hand and the item touch in quick successionContact events mark when a hand meets an item; no finer contact label is stated
An ingredient that changes shapeA cut piece is a different object from the whole onePer-frame instance masks rather than masks propagated between frames
Wet board, steel counterDepth degrades on wet and reflective surfacesPer-frame depth confidence maps; reject if more than 35% of depth pixels in the manipulation region are invalid
Break down, cut and cook as separate stepsIt is unclear where one step ends and the next beginsAction segments, one per step as written in the spec; no accuracy figure for step labels is stated
Quality of a cutWhether a piece is even or a cut is clean is a visual judgementNot stated: no thickness, uniformity or evenness label is listed
State of the foodDoneness and temperatureNot stated: no food-state label is listed, and heat is not sensed
Cutting forceHow hard the blade is pressedNot stated: no force, tactile or thermal stream is listed

Stream map

Streams for chef knife work and what the site states about an MCAP file

Left to right: the stream or item, what the site states about it, and what the site states about it in MCAP. “Not stated” means the site does not say.
Stream or itemWhat the site statesIn an MCAP delivery, as the site states it
Stereo pairDual fisheye, 1080p per eye, 62 mm baseline (dataset 06); for this row, agreed in the skill specDataset 06 is stated as shipped as MCAP; channel layout not stated
Depth with confidence mapsRecovered from the calibrated stereo pair; confidence map masks low-confidence regionsNot stated
Hand pose21 joints per hand with a per-joint visibility flagChannel layout and joint order not stated
Wrist viewsWrist cameras available alongside the head cameraNot stated
IMU9-axis (dataset 06); 6-DoF at 200 Hz (hard-environments stack)Channel layout and rate not stated
Contact events, action segmentsListed in the site’s episode schema; action segments carry start_ns, end_ns, verb and nounNot stated
CalibrationMeasured every session and shipped with the episodePlacement in the file not stated
Topic names, schemas, encoding, ratesNot statedAgreed in the skill spec

Spec checklist

What goes in the skill spec for a chef knife work collection?

Decisions the site leaves to the written spec.
Spec itemQuestion to settleHow it is settled
ScopeA single cut, a chain of cuts, or a whole step on a lineWritten in the spec; proven on a Spec Pilot
Items and toolsWhich produce and proteins, which tools, and how they beginNamed in the spec
Step boundariesWhere each action segment starts and endsVerb and noun per step
Stereo and streamsWhether a stereo pair is part of the capture, and which streams are deliveredAgreed in the spec; confirmed on a Spec Pilot
Accept criteriaWhat counts as a finished step, and what is a rejectWritten before capture; QA checks against it
Delivery formatMCAP alone or beside another format; topics, schemas and rates if MCAPMCAP, RLDS, LeRobot, WebDataset, HDF5, zarr or Rerun .rrd, or raw streams; MCAP details agreed in the spec
Privacy and licenseHandling of anyone, screens and labels in frame; ownershipBlurring as stated above; perpetual, buyer-owned, non-exclusive by default

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
Depth confidence mapsDeliveredPer frame for kitchens, to mask low-confidence regions
Consent and anonymization recordsDeliveredConsent artefact IDs per episode; anonymization record shipped with the data
FormatsDeliveredMCAP, RLDS, LeRobot, WebDataset, HDF5, zarr, Rerun .rrd or raw streams with metadata
Knife type, cut, ingredient and dishNot statedAgreed in the skill spec
Cut size, knife angle and toleranceNot statedNo such figure is listed
Cut-quality and food-state labelsNot statedNo such label is listed
MCAP topics, schemas and ratesNot statedAgreed in the skill spec
Number of sites, episodes and hoursNot statedSet per program in the skill spec

Example footage

What does egocentric stereo footage look like?

Two real clips already published on this site. The first is a head-mounted stereo capture with hands in frame, shown as a format and rig example. The second is a home-kitchen clip of an operator browning mince at a gas range, listed in the site’s Kitchen & food prep set. Neither shows a professional line or chef knife work.

egocentric_stereo_sample.mp4 · head-mounted stereo · hands in frame · 1920 × 1080 web preview · format and rig example, not a knife work clip
sample-dinner-cooking.mp4 · Kitchen & food prep · home kitchen, stirring in a skillet · 640 × 360 web preview · not a professional line or knife-work clip

No chef knife work clip is published on the site, so these are labelled as format and rig examples. Request a sample to see delivered files. See stereo sample page, sample data outputs, request an egocentric data sample.

FAQ

Stereo knife skills video, answered.

What is stereo chef knife work data?
First-person video from two calibrated cameras of a paid, consented contributor breaking down produce and proteins with precise knife cuts and cooking dishes on a professional line, captured to a written skill spec, with hand pose and sensor streams alongside.
Which catalog row covers it, and what is its status?
One row in Restaurant & Commercial Kitchen: "Culinary Chef Work / Knife Work" ("Break down produce and proteins with precise knife cuts and cook dishes on a professional line."). It is rated Hard and Active collection. This page claims no volume, hours or delivered clips.
Is the row captured in stereo?
The catalog lists Phone (head-mounted) / multiview rig. Stereo is not stated for the row; it reaches it through the general stereo offer, agreed in the skill spec and confirmed on a Spec Pilot.
Can it be delivered as MCAP, and what is in the file?
Yes. Firsthand offers MCAP delivery, and dataset 06 ships as MCAP. Topic names, message schemas and rates are Not stated on the site and are agreed in the skill spec.
Does the site name knife types, cuts, cut sizes or cut-quality labels?
No. The site states no knife type, cut, ingredient, dish, cut size, knife angle or tolerance for this row, and lists no cut-quality, thickness or food-state label. Those details are agreed in the skill spec.
How is this different from the egocentric chef knife page?
Both cover the same single catalog row. The egocentric page covers the first-person side; this page covers the stereo and MCAP side, where stereo is a spec-level choice rather than part of the row.
Is there an off-the-shelf chef knife dataset?
Not stated. This row is a custom collection, at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs. Only the stereo head rig, dataset 06, is stated as MCAP.
How fast can a fast, on-demand chef knife dataset arrive, and who owns it?
A median of a median of 48 hours from signed spec to first validated batch where coverage is deep, and three to four weeks where it is thin; the range is confirmed when the spec is quoted. The buyer owns the data under a perpetual, non-exclusive commercial license by default.

Sources

Where are the MCAP container, the formats and the public kitchen dataset documented?

Request a custom stereo chef knife work dataset.

Tell us what one step is, the produce, proteins and tools and their start state, whether stereo is part of the capture, and the streams and format you want, MCAP included. We will quote reach, timeline and price against a written spec.