What equipment do I need?

A camera that shoots slow motion and a tripod. Any phone from the last few years does 120 frames per second. A treadmill is convenient but not required, a ten-metre runway works, as long as the camera stays still.

Where does the footage go?

Nowhere. The video is processed in your own browser and what reaches the server is coordinates: twenty-six points per frame. No image and no face can be reconstructed from that. The records you build up stay on your machine too.

Then why is there a server at all?

Because everything after the points: filtering, finding stride cycles, the measurements, the quality checks, the report: is the same code that runs for practices with a local installation. That is on purpose: one product, not two that resemble each other and drift apart within a year.

Contact events are found by three independent methods and only count when at least two agree. Against a simulated runner with known ground truth, initial contact lands within zero frames and toe-off within about two. What that means on real footage belongs in a validation study rather than on a sales page; the protocol for one is written and published in the evidence section.

Can I use it on more than one computer?

The subscription covers the practice, so everyone who works there. The records, though, live in the browser you work in, so a second machine starts empty. There is a single-file export to carry them across. If you need genuinely shared records, the local installation on a practice server is the route.

Is this a medical device?

No, and that is built into the product rather than written into a disclaimer. It measures and it assesses the quality of that measurement. It does not connect those values to a condition and it does not propose a treatment. The reasoning behind that classification is a public document you can read.

Which languages does the report come in?

Dutch, English, German and Spanish, chosen per report, so a German-speaking patient gets a German report from the same measurement. The German and Spanish versions are translated but not yet reviewed by a native speaker with domain knowledge, and those reports say so on the page.

Markerless running gait analysis, from a camera you already own

Film in slow motion on a phone or a camera. A few minutes later you have twenty-six body points per frame, twenty-plus measured values with the spread across stride cycles, and a report you can put in a record.

Run your first analysis freeFull report. No watermark, no card details.

26: body points per frame, heel and toe included
20+: measured values, each with its spread across cycles
3: independent methods agree on every contact event
0: frames of footage leave your building

What this is

Video gait analysis usually means a clinician stepping through a clip frame by frame and judging it by eye. That works, and it is hard to repeat: the same clip two months later does not get judged the same way, and a colleague judges it differently again. This tool does the measuring in numbers so that the judgement stays yours and the measurement becomes reproducible.

It is markerless. Nobody sticks reflective dots on a runner, which matters less for accuracy than it does for whether the analysis actually gets done: a measurement that costs fifteen minutes of preparation gets skipped on a busy afternoon.

There is no model in here that tells you what is wrong. The report contains the measurement and an assessment of how good that measurement was. What it means for this runner is your profession, and that boundary is deliberate, see below.

How it runs

1
Record
Camera on a tripod, square to the treadmill or the runway, at hip height. Slow motion on: 60 frames per second minimum, 120 is better. Twenty seconds is plenty. There is a one-page recording protocol you can tape to the wall.
2
Load
Drop the clip into the page. It is decoded in your own browser, and the pose model runs there too. The footage does not cross the internet, because it has nowhere to go.
3
Measure
Points are filtered, stride cycles are found, and each measurement is computed per cycle and averaged. You get the spread as well as the value, because an average of three inconsistent strides is a different thing from the same average over twenty consistent ones.
4
Judge
The report opens by saying whether the recording was good enough. Frame rate too low, camera off-axis, a leg out of frame: that goes on page one, not in a footnote. A measurement you cannot trust is worse than no measurement.

What gets measured

Every value comes with the number of stride cycles it was computed over and the spread between them. Left and right are reported separately, because that is usually the point.

Spatiotemporal

Cadence
Ground contact time and flight time, left and right separately
Duty factor
Vertical oscillation of the pelvis
Speed and step length

Sagittal plane

Foot strike angle, and the strike pattern that follows from it
Shank angle at initial contact
Overstride: horizontal ankle-to-pelvis distance at contact
Knee flexion at contact, peak flexion in stance and in swing
Hip extension at toe-off
Trunk lean, split into lean from the ankle and flexion at the hip
Head position relative to the trunk

Frontal plane

Contralateral pelvic drop, measured at midstance
Peak hip adduction
Knee valgus (frontal plane projection angle)
Foot crossing the midline

Symmetry

Left-right difference in contact time
Left-right difference in peak stance knee flexion
Left-right difference in pelvic drop

What it will not do

It does not predict injury. Not as a caveat, but as a design decision: this is a measuring instrument, and it stays one. The report makes no statement about what this runner’s values mean and cites no study as though it applied to them. That is why it is not a medical device and does not carry a CE mark as one. Software that steers treatment is a different product with a different regulatory route, and pretending otherwise on a sales page is how that route starts badly.

There is a reference library covering what the literature reports about each measure. It sits apart from the measurement and contains none of your runner’s values. You read it the way you read a textbook.

Two-dimensional video has limits no software removes. Rotation about the long axis of the leg is not recoverable. Foot and ankle are screening quality on video, not measurement quality. Absolute contact time from video underestimates by roughly five to ten per cent; the left-right difference and the change over time are unaffected. All of this is printed in the report as well.

What it costs

€ 129 a month for the whole practice, unlimited, or from € 20 per analysis on a bundle. The subscription wins from about 7 analyses a month, and the pricing page does that sum for you. Bundles do not expire.

See pricing

What people ask

What equipment do I need?: A camera that shoots slow motion and a tripod. Any phone from the last few years does 120 frames per second. A treadmill is convenient but not required, a ten-metre runway works, as long as the camera stays still.
Where does the footage go?: Nowhere. The video is processed in your own browser and what reaches the server is coordinates: twenty-six points per frame. No image and no face can be reconstructed from that. The records you build up stay on your machine too.
Then why is there a server at all?: Because everything after the points: filtering, finding stride cycles, the measurements, the quality checks, the report: is the same code that runs for practices with a local installation. That is on purpose: one product, not two that resemble each other and drift apart within a year.
How accurate is it?: Contact events are found by three independent methods and only count when at least two agree. Against a simulated runner with known ground truth, initial contact lands within zero frames and toe-off within about two. What that means on real footage belongs in a validation study rather than on a sales page; the protocol for one is written and published in the evidence section.
Can I use it on more than one computer?: The subscription covers the practice, so everyone who works there. The records, though, live in the browser you work in, so a second machine starts empty. There is a single-file export to carry them across. If you need genuinely shared records, the local installation on a practice server is the route.
Is this a medical device?: No, and that is built into the product rather than written into a disclaimer. It measures and it assesses the quality of that measurement. It does not connect those values to a condition and it does not propose a treatment. The reasoning behind that classification is a public document you can read.
Which languages does the report come in?: Dutch, English, German and Spanish, chosen per report, so a German-speaking patient gets a German report from the same measurement. The German and Spanish versions are translated but not yet reviewed by a native speaker with domain knowledge, and those reports say so on the page.

Read further

Running gait analysis for physiotherapists
Measure the running pattern from a slow-motion clip, put it in the notes, and at review see whether anything actually changed. From € 20 an analysis.
Running gait analysis in sport and exercise medicine
What 2D video can and cannot measure, with the uncertainty stated. For anyone deciding whether a gait measurement will carry what gets built on it.
For podiatrists: what video gait analysis does not measure
Honest answer: foot and ankle are the weakest part on 2D video. What this adds is what happens above the foot, and what changes when the orthosis goes in.
Running gait analysis for coaches
Cadence, overstride and contact time as numbers, so that across a season you can show what changed instead of asserting it. From € 20 an analysis.
For running stores: not the gait analysis you already do
This does not tell you which shoe fits. It does show what measurably changes in the running pattern between two shoes: which is a service you can sell on its own.
Pricing for running gait analysis
Unlimited for € 129 a month, or from € 20 an analysis on a bundle. Bundles do not expire. First analysis free.
Evidence: how it measures and what has been tested
Which model sits underneath, how gait events are established, what has and has not been validated, and why this is not a medical device.
Recording protocol for running gait analysis
Camera position, frame rate, clothing and duration. The recording determines the measurement more than the software does, and this is what has to be right.
Compared: camera, IMU sensors, a motion lab or the naked eye
What camera-based analysis can and cannot do against IMU sensors, a gait lab and assessment by eye. Including the places where the others win.
Try it: the first analysis is free
One complete analysis with the report, no watermark and no card details. You need a twenty-second slow-motion clip and nothing else.
Bike fitting with the camera you already own
Measure your client on the trainer from an ordinary clip. Joint angles per crank revolution, adjustments in millimetres, a report for the file. From € 28 an analysis.
Overpronation: what it is and what it does not predict
What pronation is, why sorting runners into neutral, over- and underpronating says less than long assumed, and what you can usefully look at instead.