How to cut a dance routine down to the moves children can learn

Last updated 2026-09-22 · 8 min read

What is the short version?

You have a performance video and a school event to fill. The routine is too long and parts of it are beyond a nine-year-old, but cutting those parts out naively produces a video that teaches an impossible transition, which is worse than leaving it too hard.

1. Tell Vizard Agent who is learning it and how long it must run.
2. Let it rate the movement rather than cutting on the music.
3. Ask it to check the joins where the difficult sections were removed.

What do you need before you start?

The video and the constraints around it. Vizard Agent judges difficulty from the movement itself, but knowing the age group and the length of the performance slot turns a vague instruction to make it simpler into something it can measure the finished cut against.

• The video. The full routine as performed.
• Who is learning it. Age, and how much rehearsal time.
• The exact length. Events run to a timetable.
• Anything that must stay. An entrance, a formation, an ending.
• Whether the music must stay whole. It usually does.

What do you type into Vizard Agent?

Say what the video is for. A school performance is a different brief from a highlight reel, and knowing that children will be copying the video frame by frame changes what counts as a good cut in a way no amount of technical instruction would.

Prompt

Variants worth knowing:

• "So they can learn it from this video." The actual purpose.
• "Fourth-graders." The difficulty bar, concretely.
• "Exactly three and a half minutes." The slot, to the second.

What does Vizard Agent actually do?

Here is the Vizard Agent sequence on a performance video re-cut for a primary school event. Roughly half the list is rating the movement for difficulty and the other half is repairing the damage that removing the hard sections did to the flow of the routine.

1. Samples the video and reviews the sequence of frames.
2. Checks how to assess the movement and the flow before deciding anything.
3. Rates the opening, the middle and the closing sections for their movement patterns.
4. Builds a timestamped review sheet to decide where the cuts go.
5. Maps the easy and the difficult moves to absolute timestamps.
6. Renders a version at three minutes thirty, then trims to the exact length.
7. Checks the render visually and against its waveform.
8. Re-renders with crossfades where the audio and picture joins were abrupt.
9. Finds a second where somebody walked through the shot and replaces it.
10. Finds a pose that jumps at a join and substitutes movement from either side of it.
11. Re-checks the replaced sections frame by frame before delivery.

Step five is what makes the result teachable. Cutting on the music gives you a shorter video with the same impossible moves in it; mapping difficulty to timestamps means the removals happen where the routine is hard, which is not where the phrases end.

Step ten is the step people forget. Remove eight bars and the dancer is suddenly in a different position with no transition, so the class learns a move that cannot be performed. Vizard Agent looks at the poses either side of every cut and patches the join with adjacent movement.

Step nine is the sort of thing only a frame-by-frame review catches. A parent crossing the shot for one second is invisible at playback speed and very obvious to thirty children trying to copy what happens next.

What does the result look like?

A video of the exact length the event allows, containing a routine that a class can actually learn: the difficult passages removed, the joins bridged so the movement flows, the music continuous, and no second in it where the dancer is hidden by somebody walking past.

When does this not work well?

Some routines cannot be simplified by cutting. If the difficulty is spread evenly through every bar, or if the choreography's meaning depends on a sequence you would have to break, the honest answer is a new routine rather than a shorter one.

• Difficulty spread evenly. There is nothing to remove.
• A narrative routine. The order carries meaning.
• Tightly choreographed music. Cuts break the counts.
• Very fast cutting in the source. The moves are already hard to read.
• A single continuous take with no repeats. Nothing to bridge a join with.

How do you fix a result that came back wrong?

Name the moment and say what the class cannot do at it. Vizard Agent treats every join as its own small repair, so a single awkward transition can be rebuilt without touching the rest of the routine or changing the overall length you asked for.

• "They cannot get from here to here." That join is bridged with adjacent movement.
• "This part is still too hard." That section is rated again and removed.
• "It runs four seconds long." The length is trimmed without cutting a phrase.
• "Somebody walks through the shot." That second is replaced from nearby footage.

How does Vizard Agent compare to doing it yourself?

By hand you cut the bits that look hard and hope. The video gets to the right length, the teacher plays it to the class, and the first question is how anybody is supposed to get from the end of one section to the start of the next — which is when the editing starts again.

Common questions

Can it keep the music whole? Yes. Vizard Agent cuts picture and sound so the track still runs continuously.

What if the routine must stay a certain length? Say the number and Vizard Agent works to it exactly.

Can it slow the hard parts down instead? Yes, and it often teaches better — ask Vizard Agent for a slowed practice version.

Will it mirror the video for the class? Yes. Tell Vizard Agent and the class can copy without reversing left and right.

Can it add counts on screen? Yes. Vizard Agent can add counts or step names as captions.

What age is this for? Say the age and Vizard Agent sets the difficulty bar from it.

Can it make a rehearsal version and a performance version? Yes, from the same cut, so the two match.

Does the video have to be one take? No, but a single continuous take gives Vizard Agent more to bridge joins with.

Will it keep the entrance and the ending? Yes, if you name them as parts that must stay.

Can it remove a person from the shot? Where they pass through briefly, yes — Vizard Agent replaces those seconds.

What about a formation with several dancers? It is harder, and Vizard Agent will say which joins it could not bridge.

Can it produce a vertical version for phones? Yes, though for learning a dance the wider frame is usually better.

How long does this take? Most of the time goes on rating the movement, not on rendering.

Can it suggest which parts to drop? Yes. Ask Vizard Agent for the list before it makes the cut.