FREE image generation with no sign-up, plus daily check-in credits that climb all week. No credit card, ever.

Photo to Video AI

Not a zoom across a still: the portrait your family has always known breathes, blinks and turns its head, while the room, the clothes and the light stay as the photographer left them. Nothing to install.

Examples

Stills that started moving

Two photo to video runs from single photographs. Watch the shoulders and the eyeline rather than the mouth — small motion is what makes these portraits feel like real moments again.

A 1940s portrait, repaired first
An 1800s family group

Drag to compare before and after

AI video tool

What does photo to video AI actually do?

Photo to video AI turns a single still photograph into a few seconds of footage in which the people in it move — breathing, blinking, a small turn, a smile arriving. The framing, clothing, background and lighting stay as photographed; only the people are animated.

Video tool

The version of this everyone has seen is the Ken Burns pan: a slow push across a still while nothing in the picture actually changes. It is a camera trick performed on a photograph. What happens here is the opposite — the camera does not move and the person does. It matters most on pictures where the person is no longer around to photograph again: a studio portrait from the 1890s gave its subject one expression, held for however long the exposure took, and that expression is all anyone has.

Restoration and animation are separate jobs, so if your photograph is torn or faded, run old photo restoration first.

Creative engine

What the run involves

Upload the still, ask for small motion, pick the length and resolution, and choose your engine — up to the premium video models that keep faces steady. The credit estimate updates before you generate.

1 still
Photo in
Locked camera
Default motion
Credits back
Failed runs

The camera stays still

No pan, no push, no drift. Every bit of movement in the frame belongs to the person in it, which is what separates this from a slow zoom across a photograph and why it reads as a moment rather than as an effect.

Breath before expression

The preset leads with a breath, a blink and a slight head turn — the most believable motion, and the kind that makes a still portrait feel quietly alive.

How it works

Photo to video in three steps

Ask for less, and the portrait gives you more: in photo to video, a breath, a blink and a small turn are what make someone familiar feel present again.

1

Scan or photograph the print

A phone photo of a print is fine as long as it is square-on and evenly lit. Avoid flash — the glare off a glossy print is the one thing that genuinely blocks a good result.

2

Ask for small motion

The preset asks for a breath, a blink and a few degrees of head turn — the subtle micro-motions that make a portrait feel genuinely alive.

3

Generate and download

A few seconds of video comes back with sound, ready to send to the rest of the family.

What people animate

What people bring back

Almost all of it is family. The rest is history and archives — the photo to video runs people come back for are the ones of someone they knew.

Grandparents, most of all

One studio portrait, often the only one anyone has. Seeing the two of them turn and look at each other is the moment people describe afterwards, not the technology.

The family group nobody remembers posing for

A five-person plate from the 1800s where everybody is holding still for the exposure. Colourising and animating it turns a document into a room with people in it.

A single face from a damaged plate

Faded, scratched, silvered at the edges. Repair the still first, then animate it, and a portrait that was barely readable becomes someone looking back at you.

Historical figures and archive photographs

Museums, local history projects and school lessons all use the same trick for the same reason: a moving face holds attention that a wall of stills does not.

A memorial slideshow that is not all stills

Two or three animated frames inside a slideshow change its pace completely. If you want the still version to look its best first, the AI aging filter works the other direction on the same kind of portrait.

Anniversary and birthday messages

A short clip built from a photograph the recipient has known their whole life, sent on the day. It travels better than a card.

Answers

Animation FAQ

No — that is the Ken Burns effect, where the camera moves and the photograph does not. Here the camera is locked and the subject moves: breathing, blinking, a small head turn. It is the harder half of the problem and the reason the result reads differently.

No. A text-to-video generator invents a scene from a description; photo to video AI is constrained by a photograph that already exists — the face, the clothes and the room are fixed inputs, and the only open question is how they move. That constraint is the whole reason the result looks like your photograph rather than someone who resembles it.

Yes, and often more worth it than a clean but boring picture. Repair it first for the best result, but heavy damage is not a blocker on its own.

A few seconds, on purpose: a short run keeps the small motion focused and lifelike from the first frame to the last.

Small motion keeps the familiar face convincing: the preset uses a breath, a blink and a slight head turn, so the portrait feels alive and still looks like them.

Yes, and it is worth asking them first. A clip of a real person doing something they never did is convincing to anyone who sees it out of context.

Yes — video is a much larger job than an image edit, so it costs more per run, and the estimate is shown before you generate. For faces that need to stay convincing, paid credits open the stronger video models, which hold a face with far less drift.

Seconds to a couple of minutes depending on queue depth. A still that arrives square-on and evenly lit goes through faster than one the model has to interpret first — most of the wait on a bad scan is the model deciding what the damage is.

Kavel pricing — you pay per generation, never per seat

Credits are shared across everything Kavel runs. You see the cost before each run, and a failed job is never charged.

50% OFFYour first purchase is half price
$8.3/month
$100/year billed yearly
  • 14,400 Credits Per Year
  • Up to 1,440 videos
  • Up to 2,880 images
  • No Queue, Instant Start
  • HD, Watermark-Free Downloads
  • Commercial Use License
20% OFF - Save $70.8
Popular
$24.0/month
$288/year billed yearly
  • 54,000 Credits Per Year
  • Up to 5,400 videos
  • Up to 10,800 images
  • No Queue, Instant Start
  • HD, Watermark-Free Downloads
  • Commercial Use License
40% OFF
$30.0/month
$360/year billed yearly
  • 102,000 Credits Per Year
  • Up to 10,200 videos
  • Up to 20,400 images
  • No Queue, Instant Start
  • HD, Watermark-Free Downloads
  • Commercial Use License

Free plan

  • Get 100 credits a week with check-ins
  • 20 Free Credits with login
  • Up to 2 videos
  • Up to 4 images
  • No Card, No Account
  • Free Images And 4s Videos When Signed In
  • Free Engine Every Day

Pay safely and securely with

Payments are processed by Stripe over an encrypted checkout. Your card details are handled by Stripe and are never stored by Kavel.

Bring one back

Pick the portrait everyone in the family already knows. Watching it move is the part nobody is ready for — and if a run fails, you get your credits back.