Egocentric Slide Prep and Plate Streaking Data

Egocentric slide prep and plate streaking data is first-person video of a person preparing a slide or streaking a plate, 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 works at a bench, so the frames show the hands, the slide or plate and the swab, rather than a distant view of someone standing at a counter. On this site it is a custom collection: you name the task and the setting, 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 at a staged bench. It is not robot or teleoperation data, and it is not a fixed corpus you download. This page is a different subject from the egocentric lab pipetting and sample handling page, which centres on liquid transfer and tube work and lists these two rows alongside twelve others, and from the pharmacy and lab bench page, which spans dispensing, labeling and weighing as well as lab work. The subject here is the two catalog rows that prepare a slide and streak a plate.

People reach this topic through close cousins: plate streaking robot training data, a microscope slide preparation video dataset, egocentric microbiology bench data, or custom lab-automation 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. Nothing on this page states that the video carries a culture result, a stain result or a measurement.

A published robot foundation model paper, GR00T N1, linked in the sources below, describes a data pyramid with web data and human video at the base, synthetic data in the middle and real robot trajectories at the top. First-person human video of bench work sits in that base layer. Human capture does not replace robot data, and this page does not claim that it does.

Both tasks put a lot into the fingers. The site’s tool use page says that grip changes live in finger-level detail that RGB video alone cannot resolve, and that each needs 3D hand pose tracked through reach, grasp and tool manipulation. That page is about driving and fastening tools, not swabs or slides, so this page does not claim the same findings for bench work. What carries over is the reasoning: a clip that shows a bench somewhere in frame does not carry what a policy needs. The hands, the swab, the slide or plate and the step boundaries have to be in the episode.

Breadth matters as much as volume. A published model write-up, GEN-0, also linked in the sources, puts it this way: "data quality and diversity matters more than sheer volume." A custom collection lets you aim at the step your model is missing, such as covering a slide or drawing the final streak, rather than hoping a broad collection happens to contain it. This page does not claim that any public corpus lacks bench data.

The site’s commercial task catalog lists two rows for this subject, both in the Wet Lab & Clinical Bench environment of the Healthcare & Life Sciences sector. "Prepare a slide" reads "Drop, smear, and cover a slide." "Swab + streak plate" reads "Swab a sample and streak a plate." Both are rated Hard, captured with a multiview rig plus handheld grippers, and marked Available on request. The first table below reproduces them exactly as stated.

Other catalog rows sit next to this work and are separate listings. The same environment lists tube, reagent, centrifuge and clinical rows, and the Pharmacy & Laboratory environment lists pipetting. The second table below sets them side by side, with the skill level and relation to this page for each.

The site does not describe a sample, a stain, a culture medium, a plate type or a slide type for these rows, so none is stated here. The catalog also lists no row for reading a slide under a microscope or inspecting a plate, so those would be settled in a Spec Pilot rather than assumed.

Each attempt is delivered as one episode, and bench work needs a decision up front about what one attempt is: one slide, one plate, or a short run of several. 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 drop placed on a slide and close with the cover applied, as the catalog row words the work. That is an illustration of how a spec can cut an episode, not a stated duration. Whether a plate with several streaks is one episode or several is a spec decision.

How many contributors, repeats and variations are covered is set in the spec. The site states that Firsthand does not capture at a customer’s own facility, so the bench and the setting used are agreed in the skill spec.

A slide and plate 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. The two wrist cameras give a closer view of the hands than the head camera alone, which matters where the fingers work a small item.

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 multiview rig with handheld grippers for both rows. The site describes the multiview rig as adding synchronized third-person angles when the spec asks for them, and describes the Wet Lab & Clinical Bench setup as a staged working set; it does not explain the grippers further.

The site does not state a camera count, resolution or frame rate specific to this task. 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 drop, smear and cover, or swab and streak, 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. That is the label that carries how the fingers hold a slide edge, steady a plate and move a swab. The site describes contact events as the moment the hand meets an item, and it does not say whether a swab tip or a slide 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 sample or the result. There is no stated label for a smear quality, a streak pattern, a colony, a contamination state, a stain or a pass-or-fail score, and no contact force reading. The catalog description describes the task the contributor performs, not a delivered measurement.

Several limits the site already states apply directly to bench capture, 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. 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 glass, on a transparent lid or on a wet smear in particular, so raise it when you scope the collection.

The third is blur categories. At ingest, automated detection finds faces, plates and screens and blurs them irreversibly. The site does not say whether a culture plate, a slide label or a container marking would be affected, and it does not list tube or sample labels as blur categories, so if label text in frame matters for your use, settle it in the spec before capture.

The fourth 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 slide or plate task, the spec decides what the manipulated object is: the slide, the plate, the swab, or a pair of them.

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.

This page describes bench handling of a swab, a slide and a plate and nothing beyond it. The site states no biosafety level, no sterility or aseptic-technique claim, no culture or test result, no good-laboratory-practice or other quality-system claim, and no certification for these rows, and this page does not describe one. It covers no patient-facing task. If your program needs terms about the materials used or the conditions under which a task is done, raise them when you scope it. None of this is legal advice, and the executed agreement controls.

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. Speed comes from a warm pipeline, not from cutting consent, review or anonymization. Both rows are Available on request, and the site does not say whether the Wet Lab & Clinical Bench environment counts as deep or thin coverage, so ask which timeline applies when you request the quote. 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.

After the first batch, on-demand top-ups run against the same accept criteria, so you can add steps or variations without rewriting the spec. The site does not state a turnaround for those later batches, so plan them in your quote. 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 table below separates what is delivered from what is not stated. There is no listed label for smear quality, streak pattern, colonies, contamination, stain or result, no force or tactile stream, no sample, medium or slide specification, and no biosafety, sterility, quality-system or certification claim. Nothing here claims a training outcome or a model improvement.

The tradeoff to plan for is the embodiment gap between a human hand drawing a swab across a plate and a robot platform doing the same, which retargeting or pairing with robot data has to close. Counts of contributors, hours and repeats are set in your skill spec.

Describe the step, the item, the setting and the variations 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 scoping checklist below maps the parts of a skill spec onto a slide or plate 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, because that is what lets each step become one action segment. Name the setting and the capture tier, since the catalog lists a multiview rig with handheld grippers for both rows. And name where the episode starts and stops: one slide, one plate or a short run, since the site does not set an episode length.