Egocentric Plant Staking and Deadheading Data

Egocentric plant staking and deadheading data is recorded evidence of one person supporting a plant stem with a stake and a clip, and trimming away spent growth, collected by a paid, consented contributor who performs the task written in a skill spec. Each attempt is delivered as an episode with synchronized streams and labeled action segments. It is human capture. It is not robot or teleoperation data, and it is not a fixed corpus you download.

The site states the work in two rows of its commercial task catalog. In the Greenhouse, Nursery & Garden Centre environment, the row "Stake + clip plant" reads: "Stake and clip a stem for support." The row "Trim + deadhead" reads: "Trim spent growth and deadhead." Both are Medium rows and both are Available on request. The environment is described with the examples "Greenhouse / grow-bench set: pots, trays, plants, shears, stakes, crates". This page is built on those two rows and on the neighbouring greenhouse rows quoted below, and on nothing else the site says about plant care.

The word egocentric in the title needs a plain reading. Firsthand’s general offer is first-person capture, and the catalog lists a head-mounted phone for many rows. For these rows the catalog lists something different: a multiview rig with handheld grippers. This page therefore describes the capture setup as the rows state it, and treats first-person or stereo capture as a choice made in the skill spec, not as something the site promises here.

Two rows cover it. "Stake + clip plant" and "Trim + deadhead" both sit under the Agriculture & Grounds sector in the Greenhouse, Nursery & Garden Centre environment. Each has the skill rating Medium, the capture setup "Multiview rig + handheld grippers" and the status "Available on request". The table below reproduces the two rows, and the rows beside them that this page refers to, as the catalog states them.

On the site, "Available on request" means the work is scoped and recruited to spec when a buyer asks for it. It does not mean hours are sitting in a library. This page claims no delivered hours for either row, and no off-the-shelf staking or deadheading dataset exists; the work is custom collection. The status is shown here exactly as the catalog lists it, and it is the same for every row in this environment.

The site names no plant species, no stake type, no clip type and no item count. The words that appear are plant, stem, stake, clip, spent growth and shears, and this page uses only those. How many stakes a plant gets, which stems count as spent and what a finished trim looks like are not stated. Those choices belong in the spec, and a Spec Pilot is the place to settle them.

The catalog states one capture setup for both rows: "Multiview rig + handheld grippers". The site explains it as a staged working set recorded by a multiview rig while the operator performs the task with handheld grippers, so demonstrations map onto a robot end effector. That is the description this page uses, and it is the only capture description this page makes for the rows.

The site does not state a head-mounted phone for these rows, and it states no stereo pair for them. A calibrated stereo pair is offered generally, and a head-mounted stereo rig is the instrumented tier of the general capture stack. If a program wants either for staking or deadheading, it is agreed in the skill spec. The egocentric label in the title names the general first-person offer and the query people use; it is not a claim about these rows’ rig.

The multiview setup matters for work like this because a stem, a stake and a clip are handled in a small volume while the hands move around the plant. Synchronized third-person angles show how the two hands share the work and where the clip goes on. A rig that captures them with handheld grippers in the operator’s hands also lines up with the site’s stated reason for the setup: the demonstration maps onto a robot end effector. The site gives no camera count, resolution or frame rate specific to these rows.

Each attempt is delivered as one episode, and these tasks need a decision up front about what one attempt is. For staking it could be one stem staked and clipped, a run of several plants along a bench, or the stretch from an unsupported stem to a supported one. For deadheading it could be one trim, one plant cleared of spent growth, or a pass along a tray. The site does not set an episode length, a number of plants per episode or a number of episodes, so the skill spec does.

A short episode is easier to check against a single accept criterion, such as one stem clipped to its stake. A long episode shows the order of the work: which plant is approached next, how the stem is held while the clip goes on, and when the contributor moves from staking to trimming. That is an illustration of how a spec can cut an episode, not a stated duration.

The cut matters because of one machine-checked rule. An episode is rejected if the manipulated object is out of frame for more than 15 percent of it. In staking the manipulated object may be the stem, the stake or the clip, and in deadheading it may be the plant or the piece being trimmed. The spec should say what counts as the manipulated object at each step so the rule is applied the same way across contributors.

A second rule applies to depth: an episode is also rejected if more than 35 percent of depth pixels are invalid in the manipulation region. Thin stems and leaves are the kind of subject where depth can be patchy, and the site says depth degrades on wet and reflective surfaces and ships per-frame confidence maps. It does not say how depth behaves on plants, so this page makes no accuracy claim for them.

Inside each episode, action segments divide the recording into start- and end-timestamped spans, each carrying start_ns, end_ns, verb and noun. For these tasks a segment can be one step as written in the skill spec, so a staking pass becomes a run of labeled steps, such as stake and clip, with a noun for the item, rather than one long clip. Other standard fields include hand_joints and hand_visible for 3D hand pose, contact_events, objects and instance_masks, stored at the depth resolution of 848×480 so they line up with the depth stream.

Hand pose is delivered with a per-joint visibility flag for occluded fingers. That flag is the signal that matters when fingers close around a stem, a clip or a pair of shears. The site describes contact events as the moment the hand meets an object. It does not say which objects count in a staking or deadheading episode, so the spec settles what a contact event means for your task.

The site lists no label for the outcome of the work. There is no stated support-quality, clip-placement, trim-quality or task-completion score, and no contact force reading. The catalog verbs describe what the contributor does, not a delivered measurement. A head trajectory and a map of the bench are likewise not among the listed deliverables. The site states 6-DoF IMU data at 200 Hz from head- and wrist-mounted sensors, and nothing more about motion.

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. The official documentation for three of these formats is linked in the sources below. Collections are billed per validated hour, and the license is perpetual, irrevocable and worldwide, buyer-owned and non-exclusive by default, with exclusivity available as a separate term.

Hand occlusion is the main one. In staking the fingers, the stem, the stake and the clip all occupy the same few centimetres, and the hand closes around the work. The site’s answer is hand pose with a per-joint visibility flag, so a model can tell a hidden finger from a missing one, and multiview angles that see the hands from more than one side. The spec decides which angles matter for your task.

The second is small, similar-looking objects. Stakes, clips, stems and trimmings can look alike against a bench of pots and trays, and a per-frame instance mask is only useful if the spec says what is one object. The site delivers per-frame instance masks where the spec calls for them, and it does not say how well they separate individual stems or clips, so this page claims nothing about it.

The third is reflective and wet surfaces. A grow bench is a place where water is present and surfaces can be shiny, and the site says depth degrades on wet and reflective surfaces. Per-frame depth confidence maps are shipped so low-confidence regions can be masked, and the 35 percent invalid-depth rule rejects an episode that falls below the bar. The site does not say how often this happens in a greenhouse.

The fourth is the object leaving the frame. A trimmed piece is carried away, a stem is bent out of view while it is clipped, and the manipulated object can drop out of shot. The 15 percent out-of-frame rule exists for that, and the episode is rejected rather than quietly accepted. Rejected items are recaptured at our cost, and only footage that clears the signed accept criteria is billed as a validated hour.

The catalog states them as two rows with the same rating, the same capture setup and the same status, so either choice is open. A request for one row is a normal starting point. A request for both lets a single contributor move between the two kinds of step on one bench, which is how the work tends to read in a greenhouse. The site states no preference, so the choice is made in the skill spec.

The practical difference is the verb and the noun. "Stake + clip plant" is about giving a stem support, and its nouns are the stem, the stake and the clip. "Trim + deadhead" is about removing spent growth, and its nouns are the growth and the plant. Naming those separately is what lets each step become its own action segment instead of one long label.

These tasks are useful shapes of work because they combine several hard parts of manipulation in one short sequence: holding a deformable stem, placing a second object against it, and fastening or cutting with a tool. The site rates both rows Medium, which is the same rating it gives to pruning, trellising and transplanting in the same environment. The tasks are not simple, and this page does not claim that any current robot performs them.

First-person and multiview human video of the task sits at the base of the data pyramid the glossary explains, where volume is high and the gap to a robot body is widest. Human capture does not replace robot data, and this page does not claim that it does. The staged working set with handheld grippers is the site’s own way of narrowing that gap for rows like these, because the demonstration is made with a tool that maps onto an end effector.

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. An undocumented bystander is also one of the machine-checked reject criteria. The anonymization record is delivered with the data.

The site names no grower, farm, nursery, region or city for these rows, and it lists capture at your own facility as not offered. This page therefore makes no claim that episodes are recorded at a particular greenhouse, and none that they are recorded at yours. Capture is run to your skill spec using the settings and contributors Firsthand arranges. The site states nothing specific about workers or crops visible in the background of a greenhouse, so this page makes no claim beyond the general process.

On handling safety, the site states no sharps, shears or tool-safety protocol, no contributor safety protocol and no agricultural certification for these rows, and this page does not describe one. If your program needs terms about the tools used or the conditions under which the work is done, raise them when you scope it. None of this is legal advice, and the executed agreement controls.

The table below sets this page against the neighbouring greenhouse, nursery and produce pages on this site. Each row states the page’s main subject and the catalog rows it draws on, with the status as the catalog lists it. No delivered hours are claimed for any of them.

The short version: the potting and seed tray pages cover filling pots and sowing and replanting seedlings, the pruning and trellis page covers cutting and supporting stems as a wider set, the herb bunching page covers gathering and banding herbs, the produce packing page covers packing and grading, and the plant care page covers the wider run of gardening tasks. This page treats staking and clipping a stem and trimming and deadheading as the subject, and nothing else.

From signed spec to first validated batch, the site states a median of 48 hours where coverage is deep and three to four weeks where it is thin. It does not say whether the Greenhouse, Nursery & Garden Centre environment counts as deep or thin, so ask which timeline applies when you request the quote. Speed comes from a warm pipeline, not from cutting consent, review or anonymization.

The site also states program milestones: a pilot reaches its first delivery in two to three weeks, and a standing program reaches full rate in four to six weeks. Those assume capture coverage is already in place. After the first batch, on-demand top-ups run against the same accept criteria, so you can add plants or settings without rewriting the spec. The site does not state a turnaround for those later batches, so plan them in your quote.

Where this site does not state something, this page says "Not stated" rather than filling the gap. There is no plant species, no stake type, no clip type, no item count, no task duration, no site count, no region or city, no episode count, no support-quality, trim-quality or completion label and no tool-safety guidance. Nothing here reports a robot result, a benchmark or a model improvement.

Counts of contributors, hours and tasks are set in your skill spec. A task outside the rows quoted on this page, such as watering or tying a stem in a way the catalog does not name, is settled in a Spec Pilot rather than assumed.

Describe what one staking step is, what the clip step looks like, what counts as spent growth 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, a small, fast first batch that proves the spec and the accept criteria on real footage before volume scales. The scoping checklist below maps the parts of a skill spec onto a staking and deadheading 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. Name the step and the item rather than a broad activity, because that is what lets each stake, clip and trim become one action segment. Decide whether you want the capture setup the rows state, a multiview rig with handheld grippers, or a first-person or stereo rig agreed in the spec. And name where the episode starts and stops, since the site does not set an episode length.