How do I put text behind a moving person in a video with AI?

Write it into your brief, for example “two-word titles tucked behind the hiker”. Shorty can place text behind the subject, so the person stays in front of the words, as in the ten-second Alps reel below. For a label that should travel with something moving, name it (“pin the price to the bottle”): Shorty tracks the object frame by frame, scales the graphic as it comes closer and tucks the words behind anything that crosses in front. There is no mask to draw; check the render where the subject’s edges move fast.

Key takeaways

  • Ask for text behind the subject in the brief; there is no mask to draw.
  • Name an object to pin a label to it; Shorty tracks it, scales the label and lets words pass behind the foreground.
  • Camera tracking is a different look: the shot locks on a detail while the world moves.
  • Check faces, hair, hands and title timing in the exported MP4.

Three effects people mean by “text behind a person”

The phrase covers three different looks. Say which one you want, because each is asked for differently.

Text-behind-subject effects and how to ask Shorty for them
EffectWhat you seeHow to ask
Text behind the subjectA title sits between the background and the person, so they appear in front of the words“Tuck the title behind the hiker”
A graphic that follows a moving objectA label rides with a car, product or sign for the whole move and leaves when it does“Put the model name on the red car”
Camera trackingThe shot moves so one detail, such as a face, hand or logo, stays still while the world swings around it“Camera tracking on his head”

Real edits with text behind the subject

Shorty edits from the Discovery gallery and from a comparison published in Shorty News. Each caption says what the brief asked for; in every export the subject passes in front of the words or the label stays on the moving object.

A recorded example: titles behind a hiker

In a comparison published on 18 October 2026, Shorty’s Auto director cut ten seconds of Alps footage to the clip’s own soundtrack with two-word titles tucked behind the hiker or the mountain. Auto passed every check in that comparison. The other director tested put a title across the hiker’s eye in the opening close-up, which is the kind of placement to look for in your own export.

Sources: 16 real edits side by side. One first-party comparison; results depend on your footage and brief.

Brief: “Ten seconds of the Alps cut to the clip’s own soundtrack, with two-word titles and text tucked behind the hiker or the mountain.” Edited by Shorty on Auto. Watch the film

Ask for it in the brief

Name the subject the text should sit behind, the words, and where they should appear. Keep titles short: two or three words read in the time a shot is on screen.

Make a 12-second vertical reel from my clips. On the opening wide shot, tuck the title “SUMMER ’26” behind me so I stand in front of the letters. Later, put one short white word behind me on each of two more shots. Never let text cover my face.

Use this brief

Pin a label to something that moves

Name the object and the text, such as “pin the price to the bottle”. Shorty finds the object, follows its measured position on every frame and moves the title, label, image or shape with it, so the two do not drift apart. When the object comes toward the camera, the graphic grows with it. When it leaves the frame or turns away, the graphic exits with it instead of freezing in place.

Tracked titles can also pass behind the scene: if a pole or a passer-by crosses in front of the car, the words tuck behind and come back out, still riding with it. Object tracking is built for solid things with a clear shape, such as cars, products, signs and buildings; graphics that follow people and faces use their own tracking.

Sources: Graphics that follow moving objects · Camera tracking on any subject. Shorty News release notes for these features.

What to check in the export

Watch the downloaded MP4 on a phone and pause where the text appears.

  1. The subject is fully in front of the letters, including hair, hands and anything moving fast at the edge.
  2. No title crosses a face or covers the action the shot is about.
  3. Each title stays on screen long enough to read.
  4. Tracked labels stay attached through the whole move and leave with the object.
  5. If something is off, name the shot and the fix in a revision, such as “move the title lower, behind her shoulders”.

Where Shorty fits and where it does not

Shorty fits when you want the effect inside a finished edit made from your own clips: a travel reel, a product video, a car edit or a lookbook. You describe it; there is no tracking box to draw and no keyframes to fix.

It does not give you a frame-by-frame mask or rotoscoping tool to paint edges yourself. If a shot needs that level of manual compositing, use a dedicated compositing app. Rendering an edit needs an account and credits.

Questions, answered.

Can AI put text behind a person in a video?

Yes. In Shorty, ask for the title to be tucked behind the person in your brief. The person stays in front of the words; check hair, hands and fast edges in the export.

Can the text follow a moving person or object?

Yes. Name the object and the text, and Shorty tracks it frame by frame, scales the graphic as it comes closer and removes it when the object leaves. Object tracking is built for solid things such as cars, products and signs; graphics that follow people and faces use their own tracking.

Do I need to draw a mask or tracking box?

No. Describe the effect in your brief and Shorty finds the subject or object in the shot.

What is the difference between text behind a subject and camera tracking?

Text behind a subject layers the title between the background and the person. Camera tracking moves the shot so one detail, such as a face or a logo, stays still on screen while the background swings around it.

Take this page into your chat.

Open it in ChatGPT or Claude with the page already loaded, or read the plain-Markdown version.

Your next video starts here.

Bring the footage. Tell us the idea. Shape the edit.

Put titles behind your subject