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

Apply to capture

Wood carving

Stereo Wood Carving Data

Shape and detail wood with chisels, gouges and knives into a carved piece: one catalog row, Wood carving, rated Hard, in which tools, hands and a changing piece share one close view that a single camera cannot hold. This page covers the stereo and container side of chisel and gouge carving robot data: what a calibrated stereo pair adds to stereo wood carving video, what MCAP is according to its own documentation, and what the site does not state. We run custom woodcraft data collection fast and on demand against a spec you write, as a carved piece dataset when the spec names it, delivered as MCAP, as custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs.
Catalog row
Wood carving (Hard), in Woodcraft & Carving Studio
Status
Available on request
Container
MCAP (.mcap), open source
License
Perpetual, buyer-owned

Why stereo and MCAP here

What do a stereo pair and an MCAP file add to wood carving video?

01

Depth for a tool, a piece and two hands

The site states that a stereo pair at human-eye spacing makes depth recoverable from every frame pair, so the chisel, the gouge, the knife, the piece and both hands sit in one frame. The site gives no depth accuracy figure for this row and does not state how depth behaves on a carved surface.

02

Fingers hidden behind the tool or the piece, 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 folded behind a tool or the piece is marked as hidden rather than silently filled in.

03

Shaping and detailing as separate steps

The row names two actions, shape and detail. Action segments carry start_ns, end_ns, verb and noun, so each step can be a labeled span. Which verbs and nouns are used is set in the skill spec.

04

Several streams in one container

MCAP stores multiple channels of timestamped, pre-serialized data, so a stereo pair, hand pose and IMU can share one file on one log-time axis. Which streams go in the file is set in the skill spec.

Definition

What is stereo wood carving data?

It is first-person video of a person carving wood at a bench or in a studio, recorded with two calibrated cameras, with hand pose and sensor streams alongside. People reach it through several phrasings: stereo wood carving video, chisel and gouge carving robot data, a carved piece dataset, or custom woodcraft data collection at scale. They describe one request from different angles. Stereo describes the capture, and MCAP describes the container. They are separate choices.

The site does not state that stereo is already running on this row. The catalog lists a head-mounted phone or a multiview rig as the capture setup, so a stereo pair is a capture choice agreed in the skill spec and confirmed on a Spec Pilot. The data is human capture by paid, consented contributors, not robot or teleoperation data, so using it for a specific robot still needs your own retargeting step. Firsthand is not affiliated with Foxglove or the MCAP maintainers.

Catalog row

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

One, listed in lib/commercial-catalog-data.ts under the Woodcraft & Carving Studio environment in the Creative Workshops & Crafts sector, whose examples are woodturning, wood carving, marquetry, furniture crafting and whittling. The row "Wood carving" is described as "Shape and detail wood with chisels, gouges and knives into a carved piece." and is rated Hard.

It lists the capture setup "Phone (head-mounted) / multiview rig" and is Available on request. The site states no delivered hours and no delivered stereo footage for the row, so this page claims none. The first table below reproduces the row, and the four other rows of the studio, as the catalog states them.

Wood carving

How does the Wood carving row read, and what can a stereo spec label?

The row names a material, wood, three kinds of tool, chisels, gouges and knives, two actions, shape and detail, and a result, a carved piece. The neighbouring Whittling row names a single knife and is rated Medium, while this row is rated Hard. A spec that says "carve a piece" gives one long action segment, while a spec that names each step, such as a shaping stage and a detailing stage, and says where each starts and ends, gives a sequence you can slice by step, with each segment carrying start_ns, end_ns, verb and noun.

That is all the site states. Carving style, wood species, tool types, brands and models, the finished pieces and what "detail" means in practice are Not stated, so this page makes no claim about them. Whether an episode is one pass of a tool, one stage or a whole piece from block to carved form is a spec decision, not a default.

Hard

What does "Hard" mean for this row, and why is the task demanding?

On this site, Hard is the value in the catalog’s Skill column, and it is the only meaning this page uses. The Wood carving row carries it. Four of the five Woodcraft & Carving Studio rows are Hard, and Whittling is Medium. The site publishes no rubric, scale or criteria behind the column, and this page imports none from elsewhere.

What the row text does show is what a capture has to hold. Three kinds of hand tool are named, so the tools sit in the hands for the whole task and the hands are the data. Two actions are named, shape and detail, so the work has stages. And the result is a carved piece, so the object being worked changes shape across the episode. The site states no failure rates, tolerances or skill levels for the people doing the work, and this page states none.

Difference

How is this different from the egocentric wood carving page and the other woodcraft pages?

This page is the stereo and MCAP treatment of one row, Wood carving. The egocentric wood carving page covers the Wood carving and Whittling rows together as first-person capture, with carving as a whole as its subject, and it remains the page the hard-task hub links for this row. This page is the stereo companion to it.

The stereo woodcraft and carving page covers all five studio rows as stereo capture and MCAP delivery, so wood carving is one of five there. The stereo whittling page covers the Whittling row only, with a knife and a block of wood. The stereo sculpting page covers the Sculpting row of the Art Studio, which names wood as one of three materials. Those are different rows, and none has chisels, gouges and knives shaping a carved piece as its subject. The second table below sets the pages side by side with their status as the site lists them.

Capture setup

What capture setup does the row list, and where does stereo come in?

The catalog lists "Phone (head-mounted) / multiview rig" for the row. The site does not state a stereo pair for it. Stereo applies through the site’s general stereo offer: a stereo pair is a capture choice agreed in the skill spec and confirmed on a Spec Pilot. The catalog does not say which of its two modes applies to a batch, so the setting is agreed in the spec. The site lists capture at your own facility as not offered.

The site describes two stereo setups and does not say which one an order for this row uses. The off-the-shelf stereo head rig, dataset 06, is dual fisheye cameras at 1080p per eye, a 62 mm baseline, per-unit calibration, a 9-axis IMU and hand-tracking output. The hard-environments stack is a 4K/60 head camera with a 120° lens, two 1080p/60 wrist cameras, depth at 848×480 and 30 fps, and a 200 Hz 6-DoF IMU. Camera count, resolution and frame rate for wood carving are Not stated.

Occlusion and depth

What do the hands, the tools and the piece do to visibility and depth?

In carving the hands sit between the camera and the work, and the piece can vanish behind a finger, a closed hand around a tool or the tool itself. The site states that the hands occlude the work and that fingers disappear behind objects. That is why each hand joint carries a visibility flag, why contact events mark when the hand meets an item, and why instance masks are recorded per frame.

The site states that metric depth on sub-centimetre parts is flagged rather than interpolated, and that depth degrades on wet and reflective surfaces, with per-frame confidence maps masking low-confidence regions. Whether a carved surface or a tool edge counts as such a part or surface is Not stated. A Spec Pilot checks it on real footage rather than assuming it.

Hand pose and streams

What does a stereo wood carving episode contain?

Each episode follows the Firsthand episode schema, with head and wrist RGB, depth, 3D hand pose with per-joint visibility, contact and grasp events and frame-accurate action segments. Hand pose is delivered as handJoints[2][21][3] in metric units in the head-camera frame, two hands of 21 joints each, with a visibility flag per joint. For this row, hand pose tracks how the fingers hold the tool and the piece, contact events mark when a finger or a tool meets the wood, and action segments mark each step of the carve.

The only language data the site ties to Firsthand capture is one task instruction per episode. Object labels use the terms your spec sets. Force or tactile data, and success or quality labels for a finished piece, are Not stated for this row, and this page does not offer them as deliverables. The site offers a tactile glove tier as a pilot-ready option, but it does not state that the tier has been used on woodcraft.

Not stated

What does the site not state about stereo wood carving?

Carving styles, wood species, tool types, brands and models, piece sizes, cut depths, tolerances, cycle times, error rates, item counts and what a finished carved piece means are Not stated. No carver, maker, studio, client, piece or commission is named for the row, and none is used here. They are set in the written spec, and this page claims none of them.

Safety and privacy

What does the site say about safety, consent and privacy?

Tool safety is Not stated on the site. It states no contributor safety protocol, no injury-prevention guidance and no trade certification for the studio rows, so this page describes none and gives no safety or carving instructions. Who performs the work and under what conditions is something to raise when you scope the order.

For every delivered episode the site states that faces, plates and screens are detected and irreversibly blurred at ingest, that a reviewer confirms the result before anything ships, and that footage that cannot be de-identified is discarded. The anonymization record ships with the data, an undocumented bystander is a stated reject criterion, and every episode links to a signed contributor release through a documented consent chain. This is not legal advice, and the executed agreement controls.

MCAP

What is MCAP, per its own documentation?

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.

Reader and writer libraries exist for C++, Go, Python, Rust, Swift and TypeScript, and the mcap CLI can inspect, merge and split files. The foxglove/mcap repository is MIT-licensed; that is the MCAP project licence, not the licence of a Firsthand delivery. The MCAP hub page and the glossary entry cover the format in full.

Streams

Which streams can ship in an MCAP delivery of wood carving data?

Firsthand offers MCAP delivery, and dataset 06 is stated as shipping as MCAP. For this row the site states hand pose, a stereo-derived depth stream with confidence maps, IMU, contact events, action segments, per-frame instance masks and one task instruction per episode elsewhere. What it does not state is which of those are channels in an MCAP file, how they are named, what encoding they use or at what rate. The third table below maps each stream to what the site states.

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.

Failure modes

What goes wrong when you record close hand-tool work in stereo, and what does QA reject?

Hand occlusion by the tools and the piece is the first risk, handled by the visibility flag above. Calibration is measured at the start and end of every session under a 0.28 px RMS gate, and an episode is rejected for a sync breach beyond the ceiling for more than three consecutive frames, for the manipulated object out of frame for more than 15 percent of the episode, for more than 35 percent invalid depth pixels in the manipulation region, for an undocumented bystander or for a mismatch against the named skill spec.

For this row, what counts as the manipulated object, the piece, the tool or the area being worked, is a spec decision, and it decides how the out-of-frame rule applies. The site does not describe how a piece held in the hand, a fast-moving tool or loose material on the surface is handled, so settle that on a Spec Pilot. A reject log ships with every batch, and rejected items are recaptured at the provider’s cost. The fourth table below lists each failure mode with what the site states.

Scale

How do off-the-shelf coverage and custom collection combine for wood carving?

The October 2026 off-the-shelf catalog lists ~324,800 hours across 18 datasets, of which stereo is ~64,800 hours across 9 datasets. The site names no wood carving dataset among them, so this work is custom collection, not an off-the-shelf purchase. The off-the-shelf stereo and hand-tracking video can still show the format before you write a spec.

It is 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. The row is Available on request, so the site does not say which applies, and the timeline is quoted against your spec. Collections are billed per validated hour, and rejected episodes and recaptures are never billed.

Ordering

How do you order a stereo wood carving dataset?

Describe the task and where it starts and ends: shaping and detailing wood with chisels, gouges and knives into a carved piece. Say whether an episode covers one pass, one stage or a whole piece, what counts as the object and as a reject, the setting and lighting, and the privacy rule.

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.

At a glance

What you get.

Catalog row: Wood carving
"Wood carving": "Shape and detail wood with chisels, gouges and knives into a carved piece."; Woodcraft & Carving Studio, Creative Workshops & Crafts; Hard; Available on request
Catalog capture setup
Phone (head-mounted) / multiview rig. The site states no stereo pair for this row, so a stereo pair is agreed in the skill spec through the general stereo offer
Stereo head rig (06)
Dual fisheye, 1080p per eye, 62 mm baseline, per-unit calibration, 9-axis IMU, hand-tracking output; shipped as MCAP
Hand pose
21 joints per hand with a per-joint visibility flag
Sync
Hardware-triggered onto one clock; 2.0 ms cross-stream ceiling
Carving style, wood species, tool types and brands, piece sizes, cut depth, cycle times
Not stated; agreed in the skill spec
Tool safety
Not stated beyond the privacy steps listed below; nothing here is safety guidance
MCAP topics, schemas, rates
Not stated on the site; agreed in the skill spec
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 rows

Which catalog row covers wood carving, and what is its status?

The five Woodcraft & Carving Studio rows from lib/commercial-catalog-data.ts, reproduced as the catalog states them, with the wood carving row first. None of the rows states a stereo pair.
Task (as listed)Description (as listed)EnvironmentSkillCapture setupStatus
Wood carvingShape and detail wood with chisels, gouges and knives into a carved piece.Woodcraft & Carving StudioHardPhone (head-mounted) / multiview rigAvailable on request
WhittlingCarve small figures and objects from a block of wood with a knife.Woodcraft & Carving StudioMediumPhone (head-mounted) / multiview rigAvailable on request
WoodturningMount a blank on the lathe and shape it with turning tools as it spins.Woodcraft & Carving StudioHardPhone (head-mounted) / multiview rigAvailable on request
MarquetryCut veneer pieces to pattern and glue them into a decorative inlay on a panel.Woodcraft & Carving StudioHardPhone (head-mounted) / multiview rigAvailable on request
Furniture craftingCut, join and finish timber components into furniture pieces at the bench.Woodcraft & Carving StudioHardPhone (head-mounted) / multiview rigAvailable on request

Comparison

How does this page compare with the neighbouring woodcraft pages?

The existing pages closest to this topic, set against this page by capture framing, rows covered and row status as the site lists it.
PageCaptureWhat it coversRow status as listedRelation to this page
This page: stereo wood carving dataStereo pair through the general offer; MCAP deliveryOne row: Wood carvingAvailable on request (Hard)Wood carving as the main topic, in stereo
Egocentric wood carving dataEgocentric, not stereoTwo rows: Wood carving and WhittlingAvailable on request (Hard, Medium)The same row as first-person capture, with whittling beside it; not stereo as its subject
Stereo woodcraft and carving dataStereo, agreed in the spec; MCAP focusAll five studio rowsAll five Available on request (four Hard, one Medium)The whole studio in stereo; wood carving is one of five there
Stereo whittling dataStereo, agreed in the spec; MCAP focusOne row: WhittlingAvailable on request (Medium)A different row in the same studio, with a knife only
Stereo sculpting dataStereo, agreed in the specOne row: Sculpting, in the Art StudioAvailable on request (Hard)A different row that names wood as one of three materials
Egocentric stereo data in MCAPThe stereo and MCAP hub across tasksNot applicableNot applicableGeneral stereo and MCAP treatment; not specific to woodcraft

Stream map

Streams for wood carving data 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)Dataset 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
IMU9-axis (dataset 06); 6-DoF at 200 Hz (hard-environments stack)Channel layout and rate not stated
Contact events, action segments, instance masksListed in the site episode schema; masks stored at 848×480Not stated
Task instructionOne per episodeNot stated
CalibrationMeasured every session and shipped with the episodePlacement in the file not stated
Topic names, schemas, encoding, ratesNot statedAgreed in the skill spec

Failure modes and QA

What goes wrong in wood carving capture, and what QA does about it

Left to right: the failure mode, what the site states, and what a buyer sees. “Not stated” means the site does not say.
Failure modeWhat the site statesWhat a buyer sees
Fingers hidden behind a tool or the piecePer-joint visibility flag on hand pose; wrist views and a multiview angle available in the specA hidden joint flagged as hidden, not guessed
Piece or tool out of framePer-frame instance masks; episode rejected when the object is out of frame for more than 15 percent of itReject log per batch; what the object is gets set in the spec
Small partsMetric depth on sub-centimetre parts is flagged rather than interpolatedWhether a carved detail counts: Not stated; test on a Spec Pilot
Reflective surfacesDepth degrades on wet and reflective surfaces; confidence maps mask low-confidence regionsBehaviour on a tool edge: Not stated; test on a Spec Pilot
Invalid depth in the manipulation regionEpisode rejected when more than 35 percent of depth pixels are invalidReject log per batch; confidence maps ship with depth
Bystander in frameFaces, plates and screens blurred and reviewer-confirmed; undocumented bystander rejected; footage that cannot be de-identified is discardedReject log per batch; anonymization record ships with the data
Calibration driftSolved at the start and end of every session; 0.28 px RMS gateCalibration shipped with the episode
Sync breachRejected beyond the 2.0 ms ceiling for more than three consecutive framesReject log per batch
Spec mismatchRejected for a mismatch against the named skill specReject log per batch

Stated vs not stated

What the site states about stereo capture for wood carving

Left to right: the item, what the site states, and what that means for a wood carving order. “Not stated” means the site does not say.
ItemWhat the site statesFor a wood carving order
Catalog capture setupPhone (head-mounted) / multiview rig on the rowNo stereo pair is stated for this row; a stereo pair is a spec choice through the general stereo offer
Stereo head rig (dataset 06)Dual fisheye, 1080p per eye, 62 mm baseline, per-unit calibration, 9-axis IMU, hand-tracking output; shipped as MCAPWhether an order uses it: Not stated
Hard-environments stack4K/60 head camera, 120° lens; two 1080p/60 wrist cameras; depth 848×480 at 30 fps; 200 Hz 6-DoF IMUWhether an order uses it: Not stated
Row statusWood carving is Available on requestNo volume, hours or delivered stereo footage is claimed on this page
SkillHard, as the catalog’s Skill columnNo rubric, failure rate or worker skill level is stated or claimed
Calibration and synchronizationMeasured every session, 0.28 px RMS gate; hardware-triggered, 2.0 ms cross-stream ceilingStated
Carving style, wood, tool details; piece sizesNot statedNot claimed; set in the spec
Tool safetyNot stated; no contributor safety protocol or trade certification for the studio rowsNot claimed; no safety instructions are given
Task time, error rate, piece qualityNot statedNot claimed on this page
MCAP topics, schemas, ratesNot statedAgreed in the skill spec

Skill spec

What the skill spec fixes for a stereo wood carving 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
RowWood carving, worded as the catalog words itSet per program; Available on request
Steps and orderWhat one step is: a shaping stage, a detailing stage, a change of tool or gripEpisode boundaries and action segments; set per program
Episode scopeOne pass, one stage or a whole pieceSet per program; the site does not state an episode length
Object definitionWhether the piece, the tool or the worked area counts as the objectContact events and the out-of-frame reject rule; set per program
Rig and streamsStereo pair, multiview angles, wrist views, hand pose, IMU and any othersThe channels you can read; confirm on the Spec Pilot
Setting and lightingWhere capture runs and how the work area is litSet per program; the site states no lighting setup for this row
MCAP topics and encodingWhat your reader expectsSet per program
Delivery formatHow the data is packagedMCAP alone, or beside RLDS, LeRobot, WebDataset, HDF5 or zarr
Accept criteriaWhat passes QA, including what counts as a rejectReject log per batch
PrivacyHow people, screens and signage in frame are handledFaces, plates and screens blurred at ingest; other items agreed in the spec
LicenseWho owns the datasetPerpetual, irrevocable, buyer-owned, non-exclusive by default

Example footage

What does egocentric stereo capture look like?

One real clip already published on this site, shown as a format and rig example: a head-mounted stereo capture with hands in frame. It does not show wood carving or any woodcraft task.

egocentric_stereo_sample.mp4 · head-mounted stereo · hands in frame · 1920 × 1080 web preview · rig example, not a wood carving clip

No wood carving or woodcraft clip is published on the site, so this is labelled as a format and rig example. Request a sample to see delivered files. See stereo sample page, stereo + wrist-cam example videos, request an egocentric data sample.

FAQ

Stereo wood carving data, answered.

What is stereo wood carving data?
First-person video from two calibrated cameras of a person shaping and detailing wood with chisels, gouges and knives into a carved piece, with depth and hand pose alongside.
Which catalog row covers it, and what is its status?
One, under Woodcraft & Carving Studio: "Wood carving" (Hard). It lists "Phone (head-mounted) / multiview rig" and is Available on request. This page claims no delivered hours.
Is stereo already captured on this row?
Not stated. The catalog lists a head-mounted phone or multiview rig and no stereo pair. A stereo pair is agreed in the skill spec through the general stereo offer and confirmed on a Spec Pilot.
How is this different from the egocentric wood carving page?
The egocentric page covers the Wood carving and Whittling rows together as first-person capture that is not stereo as its subject. This page is stereo capture of the Wood carving row with MCAP delivery. The stereo woodcraft and carving page covers all five studio rows, and the stereo whittling page covers the Whittling row.
Is there an off-the-shelf wood carving dataset?
No listed dataset names wood carving. Stereo wood carving data 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 wood species, tool types or safety guidance?
Not stated. The site names wood, chisels, gouges and knives in the row and lists no carving styles, wood species, tool types, piece sizes, cycle times or error rates, and gives no tool safety guidance, so none is claimed.
Can it be delivered as MCAP, and what is in the file?
Yes. Firsthand offers MCAP delivery. Topic names, message schemas and rates are Not stated on the site and are agreed in the skill spec.
Is the footage anonymized and consented, and who owns it?
Every contributor signs a written release, and each episode links to it by consent artefact ID. Faces, plates and screens are blurred irreversibly, and footage that cannot be de-identified is discarded. The license is perpetual, irrevocable and buyer-owned.

Sources

Where is MCAP documented?

Request a custom stereo wood carving dataset delivered as MCAP.

Tell us the task and where it starts and ends, whether an episode covers one pass, one stage or a whole piece, what counts as the object, the setting and lighting, the privacy rule, the streams you want in the file and whether MCAP sits beside RLDS, LeRobot or WebDataset. We will quote reach, timeline and price against a written spec, with a first validated batch proven on a Spec Pilot.