Childhood photo + selfie today
AI Hug Video With Your Younger Self
An AI hug video turns two separate photos into a short clip in which the two people move toward each other and share a real, generated hug. For the hug-your-younger-self version, Photo 1 is you today and Photo 2 is you as a child: the AI hug generator is told to keep the two ages apart and makes a 5 or 10-second MP4 in your phone browser.
- 2 photos · one person each
- Real hug motion, not a slideshow
- 5 or 10 seconds · 9:16 or 16:9
- From 4 credits
Default: Seedance 2.0 Mini · 5 s · 720p · vertical 9:16 · gentle hug · 8 credits.
Hug your younger self
1. Add two photos, one person in each
Photo 1 · you today
A clear, recent selfie; this photo also sets the light and place
Photo 2 · you as a child
A childhood photo or a phone scan of a print, cropped to just you
2. Choose the hug
3. Model, length and quality
Pick the frame now: vertical for TikTok, Reels and Shorts, wide for a slideshow or YouTube.
Add a photo of you now and one of you as a child.
Your photos are only used to make your video and never appear in the examples.
When the site watermark is on, accounts without a purchase get a marked copy. Any purchase unlocks the original.
Younger-self hugs
What an adult-and-child-self hug can look like. Your clip uses your two photos.
AI hug video examples: hugging a younger self
Three short clips of an adult and a younger version of the same person. Notice what makes them work: the two faces stay different ages, the hug is the only action, and the camera barely moves. Results from your own photos will vary with photo quality.
Hug your younger self: the still picture vs a moving AI hug
The AI hug video first spread on TikTok in late 2024, when people put two separate photos into Luma's Dream Machine and posted the moment the two people embrace. In 2025 a quieter variation took over feeds: Gemini-made Polaroid pictures of an adult holding their childhood self, often posted as a photo with a nostalgic song.
Those Polaroids are still images. If you want the moment itself — child-you running in, adult-you kneeling, the two of you holding on — you need a photo-to-video model that takes both pictures as identity references. That is what this ai hug generator does: no image-to-video of one combined picture, no template that slides two photos together.
| Gemini Polaroid picture | This AI hug video | |
|---|---|---|
| What you get | One still image, usually in a white Polaroid border | A 5 or 10-second MP4 of the two of you moving into a hug |
| What you upload | A childhood photo, a current photo and a written prompt | The same two photos; the hug directions are built in |
| Movement | None (some people pan across it in an editor) | Generated walk-in or turn, arms, embrace and a final smile |
| Posting | Photo post or carousel with a song | Reel, TikTok or Short; add the song in the app |
Two photos, a real hug: what this AI hug generator does
Many hug templates only place two photos side by side and fade between them. Here the model receives Photo 1 and Photo 2 as two separate identity references and generates new frames: the two people turn or step toward each other, raise their arms, hold the hug and ease apart with a small smile. One steady eye-level shot, no cuts or zooms.
The instructions also tell the model what not to add: no kissing, no talking, no angel wings or heaven backdrop, no captions or logos, and no third person. Faces are kept slightly offset at three-quarter angles so one face does not pass through the other — the most common way a two person hug AI video goes wrong.
It is still a generated video. Hands, hair and the edges of clothes can drift, and the child will not always be exactly the right height. It makes one hug between two people; group hugs and pets are not what it is built for.
Getting your childhood photo ready
The childhood picture is almost always the weaker of the two photos. A few minutes here does more for the result than any setting.
| Your childhood photo | What to do before uploading |
|---|---|
| A printed photo in an album | Take it out of the sleeve, lay it flat by a window and photograph it straight on with your phone. Avoid the ceiling light's glare. Phone scan apps that flatten and crop the edges work well too. |
| You are with siblings or parents | Crop to just the child you were, head and shoulders at least. Two children in Photo 2 make the model guess who is who. |
| A tiny face in a wide shot | Crop closer, but keep the image at least 300 px on each side. If cropping leaves less, use a different photo. |
| Faded, yellowed or black-and-white | It can work as it is; the model will usually render child-you in colour. If the face is blurry, a gentle restore helps more than heavy colourising, which can change the face. |
| An iPhone HEIC or Live Photo | Pick it from Photos in Safari; it is normally converted to JPG on upload. If not, take a screenshot of it and use that. |
| Photo 1, you today | One person, face in soft light, no sunglasses or heavy filter. Photo 1 also decides the lighting and kind of place, so a plain indoor shot gives a calmer scene. |
Settings for a younger-self hug
Two hug styles, two lengths and two frames. This is how they suit a childhood-self clip.
| Choice | Pick it when |
|---|---|
| Gentle hug | You want adult-you to simply hold child-you: quiet, close, the calm version most younger-self posts use. |
| Reunion hug | You want the two of you to notice each other first and step in — closer to the 'finally meeting' feel. |
| 5 seconds | A first try or a loop. Less time on screen means less chance of a face drifting. |
| 10 seconds | You want the hug held long enough to carry a song lyric or a caption. |
| Vertical 9:16 | TikTok, Reels, Shorts and stories. |
| Wide 16:9 | A birthday slideshow, a family screen or YouTube. |
| Ambient sound | You want room tone under the clip. It never adds music or voices, so most people leave it off and add a song when posting. |
Seedance 2.0 Mini is the default and the cheapest. Seedance 2.0 Pro costs more and adds 1080p; try Mini first and switch only if the faces need more detail.
When adult you and child you start to look alike
Two photos of the same person are the hardest pair for any AI hug video: the model sees one face twice. These are the usual symptoms and what to change.
| What you see | What to change |
|---|---|
| Child-you has your adult face, or both look the same age | Use a childhood photo where your face is clearly a child's: round cheeks, front-facing, around age 4–10. Teen photos are much easier to blend with your adult face. |
| The child is as tall as you | Use a childhood photo that shows more of the body, not only the face, and choose the gentle hug, which keeps both standing close. |
| Faces press together or pass through each other | Use 5 seconds and the gentle hug. Faces that look straight at the camera in both photos blend more; a slight turn in one photo helps. |
| Clothes or hair change halfway | Crop out other people and busy backgrounds from both photos; the model copies clothing from each photo. |
| The scene looks dark or odd | Photo 1 sets the light and place. Swap in a brighter selfie taken indoors. |
A failed generation refunds its credits automatically. A finished video that you simply do not like uses its credits; change one thing above before trying again.
The hug, second by second (and the exact prompt)
There is nothing to write: the directions are built in. For a 5-second gentle hug the model is asked for this timing; a 10-second clip stretches each part (0–3, 3–8, 8–10 s).
- 0–1.5 s · standing close, both turn gently toward each other
- 1.5–4 s · arms come up slowly and settle into a calm hug
- 4–5 s · they hold it, breathing quietly, with a small smile
Full AI hug video prompt (5 s, gentle, silent)
Create a 5-second natural, affectionate hug video using exactly two uploaded identity-reference images. Image 1 defines person A; image 2 defines person B. Keep each person's own facial features, age, skin tone, hair, body proportions and clothing distinct throughout. Do not merge faces, swap identities or invent a third person. Place both people together in a simple, softly lit setting inspired by image 1. Match the lighting without changing either identity. Show both from at least the waist up with their arms visible. Use one steady eye-level medium shot, no cuts, zooms or photo transitions. Seconds 0–1.5: already standing close, they turn gently toward each other. Seconds 1.5–4: both slowly raise their arms and settle into a calm, comforting hug. Seconds 4–5: hold the hug with quiet breathing and a small smile; no dramatic gestures. Keep faces slightly offset at three-quarter angles so neither face passes through the other. Hands rest naturally on the shoulders or upper back, with no extra fingers, limbs or intersecting arms. No kissing, talking, fantasy transformations, angel wings, captions, logos or watermarks. Silent video, no music and no speech.
This prompt describes a two-image reference video, so it only fits tools that accept two separate identity photos. It is shown so you know exactly what the model is told.
Is the AI hug video generator free? Prices, watermark and refunds
Each video uses credits; there is no subscription. The table lists every model, length and resolution. Hug style, frame and sound do not change the price.
| Model | Length | Resolution | Credits |
|---|---|---|---|
| Seedance 2.0 Mini | 5 s | 480p | 4 |
| Seedance 2.0 Mini | 5 s | 720p | 8 |
| Seedance 2.0 Mini | 10 s | 480p | 7 |
| Seedance 2.0 Mini | 10 s | 720p | 15 |
| Seedance 2.0 Pro | 5 s | 480p | 13 |
| Seedance 2.0 Pro | 5 s | 720p | 27 |
| Seedance 2.0 Pro | 5 s | 1080p | 68 |
| Seedance 2.0 Pro | 10 s | 480p | 26 |
| Seedance 2.0 Pro | 10 s | 720p | 54 |
| Seedance 2.0 Pro | 10 s | 1080p | 135 |
Prices checked October 27, 2026. The number on the Generate button is what is charged.
| Before you start | What you get |
|---|---|
| Login | You need a free account to generate, so your photos and videos stay private to you. There is no no-sign-up mode. |
| Free credits | New accounts currently get 1 free credits, valid for 3 days. The cheapest hug video costs 4 credits. That alone does not cover a hug video. |
| Smallest pack that covers a video | With no credits left, the smallest pack that covers a 4-credit hug is Starter: $4.99 for 10 credits. |
| Billing | One-time credit packs that never expire. No recurring subscription. |
| Watermark | Videos made without any purchase carry a moving site watermark. Any purchase unlocks the clean original of videos you already made, without generating again. |
| Download | The finished MP4 downloads from the result card and stays in My videos. On a phone it saves to Files or Downloads first; move it to Photos or Gallery from there. |
| Failed videos | If a generation fails, its credits are refunded automatically. A finished video is not remade for free. |
Make it and post it
Four steps from two photos to a clip on your feed.
Add the two photos
Photo 1 is you today, Photo 2 is you as a child, each cropped to one person.
Choose the hug and frame
Gentle or reunion, 5 or 10 seconds, vertical for TikTok and Reels or wide for a slideshow.
Generate and save the MP4
Generate, wait a few minutes, then download the MP4 from the result card.
Add the song and the frame in the app
Pick a nostalgic track from TikTok's or Instagram's own sound library, add a Polaroid-style frame or a 'then and now' caption in the editor, and turn on the AI-generated label.
Want the Polaroid look from the photo trend? The generator never draws frames or captions, which keeps the hug clean. Add the white border in CapCut, Instagram or any video editor after downloading.
AI hug video with your younger self: FAQ
Can I make an AI hug video with my younger self from one childhood photo and one selfie?
Yes. Put the recent selfie in Photo 1 and the childhood photo in Photo 2, each showing one person. The AI hug video generator treats them as two different people and is told to keep each one's own age and features, then generates the two of you hugging.
Is this the same as the Gemini 'Hug Your Younger Self' Polaroid?
It is the same idea, but a video instead of a still. The Gemini trend makes one image of you holding your child self. This page makes a 5 or 10-second clip in which you move into the hug. You can add the Polaroid frame afterwards in a video editor.
My childhood photo is old, blurry or black-and-white. Will it work?
Usually, yes, as long as the face is visible and the image is at least 300 px on each side. Photograph the print straight on without glare and crop to just you. Child-you will normally appear in colour; a light restore of a very blurry face helps more than strong colourising.
Why does child me end up with my adult face?
Because both photos show the same person, the model can blend the two ages. A clearly young childhood photo (roughly 4–10 years old), a short 5-second gentle hug and a slight head turn in one of the photos reduce it.
My only childhood photo includes my brother. What do I do?
Crop the photo so only you are in it before uploading. Each slot must contain exactly one person, otherwise the model has to guess who should hug.
Can the video have music or end in a Polaroid frame?
The generator makes the hug only: silent by default, or with soft ambient sound, never music, captions or frames. Add a song from TikTok's or Instagram's sound library and a Polaroid-style border in the app's editor when you post.
Is this AI hug video generator free, and is there a watermark?
You need a free account to generate, so your photos and videos stay private to you. There is no no-sign-up mode. New accounts currently get 1 free credits, valid for 3 days. The cheapest hug video costs 4 credits. That alone does not cover a hug video. With no credits left, the smallest pack that covers a 4-credit hug is Starter: $4.99 for 10 credits. Videos made without any purchase carry a moving site watermark. Any purchase unlocks the clean original of videos you already made, without generating again.
Does it work on my phone without an app?
Yes. It runs in Safari or Chrome on iPhone and Android. You sign in, add the two photos from your camera roll, and download the MP4 when it is ready.