Two photos. Across the screen. Together.
Pull Out of TV AI Trend
The pull out of TV AI trend is a video effect where someone reaches through a television, steps into the room and shares a hug. Add two photos to recreate the TREND DA TELEVISÃO moment.
10 seconds · 9:16 MP4 · From 17 credits · Default 720p: 36 credits
Who is pulling whom out of the TV?
1. Give each person a role
Photo 1 · In the room
Photo 2 · Inside the TV
One person per photo; show the face, arms and as much of the body as possible. JPG, PNG or WebP · 10 MB max each · 300 px minimum on the short side.
2. Choose the reunion
3. Model and video quality
Fixed at 10 seconds and 9:16. A wide view of the room keeps the TV, cabinet and both people in the vertical frame.
Optional room or lighting detail. The preset already includes the fist bump, hand pull, TV exit and final hug.
Add the person in the room first, then the person coming out of the TV.
Saved to your account with no automatic public posting. Media uses unlisted links; anyone with a direct file link can open it.
Unpaid accounts get a moving site watermark. Any completed pack or plan purchase unlocks the clean download.
Watch the TV-to-room moment
Look for the hand contact, the step over the cabinet and the final hug. Audio varies by clip. Your photos and settings change the result.
What is the pull out of TV AI trend?
The pull out of TV AI trend turns two separate portraits into a moving reunion: one person waits beside a television, the other appears on its screen, they make hand contact, and the person on screen steps into the room for a hug. It is an AI video effect, not a smart-TV feature or a still-photo filter.
The name is associated with TREND DA TELEVISÃO, a Portuguese TV-effect template name, and tutorials showing a fist bump, purple screen glow and a final embrace. This tool recreates that sequence from your two photos with a timed prompt. It does not import an editing-app template or require footage of either person beside a real TV.
A French-language tutorial posted on October 22, 2026 documents the TV pullout sequence. The exact original app and a single definitive song have not been independently established here. TV trend background: October 22 tutorial
Which photo goes in the room, and which goes on the TV?
Photo order controls the roles: photo 1 stays in the room; photo 2 appears on the screen and climbs out. Use two separate images, even if the original is a single group photo. Similar lighting and visible arms give the model less missing detail to invent.
| Your starting photos | Prepare them this way |
|---|---|
| Two separate portraits | Put the room person in slot 1 and the TV person in slot 2. Keep one unobstructed face per image. Waist-up or full-body photos give more information for reaching, stepping and hugging than tight selfies. |
| Only one photo together | Make two crops before uploading: one crop for each person, with enough shoulder and body detail to identify them. Each crop must still be at least 300 px on its short side. Uploading the group shot twice can confuse the roles. |
| Different rooms, same reunion | The photos do not need matching backgrounds or a television. The model constructs a living room and places the two identities in it. Avoid contradictory room instructions; exact original backgrounds are not preserved. |
| A wide TV in a vertical clip | The output is always 9:16. The prompt uses a wide enough camera view to include both people and the cabinet inside that frame. Do not ask for a close-up while the person is climbing out. |
TV pullout videos: check the handoff and the landing
Watch the face inside the screen, the hand contact and the moment both feet reach the room. These clips show variations of the TV-exit effect; camera placement, timing and sound can vary. A generated result may skip an action or distort a hand.
A living-room reunion
10.1 s · Portrait · Use player controls for sound
A hand through the screen
10 s · Portrait · Use player controls for sound
How to make the pull out of TV AI trend today
- 01
Upload the room person, then the TV person
Choose two clear photos with one person in each. Slot 1 is the person waiting in the room; slot 2 is the person who will step out of the screen. You do not need a real television in either photo.
- 02
Pick a style and check the credits
Choose Couple or Friends. Both include the TV exit and a hug. Mini defaults to 720p for 36 credits; select 480p for 17 credits. Pro also offers 1080p. Ambient sound is optional and does not change the price.
- 03
Generate, inspect and download
Sign in, confirm photo permission and start the 10-second video. Check the two faces, hand contact and landing before saving the MP4. The task stays in Creations if you close this page.
Why do faces swap, hands merge or legs pass through the cabinet?
The screen exit changes the person from a flat TV image into a body in the room. That transition, followed by two overlapping faces and arms, is difficult for a video model. Use clear portraits with comparable face angles, remove extra characters from the photos and keep the note to one simple setting detail.
The preset asks for separate identities, one visible hand contact, feet clearing the cabinet and a steady camera. Those instructions reduce ambiguity; they do not guarantee the anatomy or likeness. A completed clip with a visual defect still uses credits, and another attempt is charged again. More pixels alone will not fix a confused pose.
This preset is built for two people. Pets need different movement and are not a supported version here. For a family tribute, choose a clear photo and get appropriate permission; the clip is an imagined scene. If you only want a reunion without the screen exit, try the simpler AI Hug tool.
Make a simpler AI hug ↗Credits, free access, sound and watermark
Every 10-second video uses credits. Couple and Friends cost the same, and ambient sound adds no charge. The default is Mini 720p for 36 credits. The table shows every available model and resolution.
| Model / resolution | 10 seconds |
|---|---|
| Seedance 2.0 Mini · 480p | 17 credits · ≈$0.47 |
| Seedance 2.0 Mini · 720p | 36 credits · ≈$1.00 |
| Seedance 2.0 Pro · 480p | 60 credits · ≈$1.66 |
| Seedance 2.0 Pro · 720p | 129 credits · ≈$3.58 |
| Seedance 2.0 Pro · 1080p | 320 credits · ≈$8.87 |
Dollar equivalents use Studio monthly: $49.90 for 1,800 credits ($0.02772 per credit). They are not separate pay-per-video checkout prices. A new generation or a manual retry costs credits again.
| Access or purchase | What you get |
|---|---|
| Free account | Eligible new accounts may receive up to 5 credits, valid for 7 days. That covers 0 complete TV pullout videos: the minimum is 17 credits. There is no daily free-render allowance. Browsing needs no account; generation requires sign-in. |
| One-time Trial pack | $5 buys 24 credits without a subscription. This covers 1 Mini video at 480p for 17 credits, leaving 7. It does not cover the default 720p option, which needs 36 credits; choose 480p or buy a larger pack or plan. |
| Monthly subscriptions | Creator: $29.90 / 800 credits. Studio: $49.90 / 1,800 credits. Monthly credits expire at the paid billing period’s end; one-time pack credits last 365 days. |
| Watermark and downloads | Unpaid-account videos carry the moving site watermark. Any completed pack or subscription purchase unlocks the clean MP4. There is no separate download charge. |
| Failed video or unwanted result | A failed generation returns its credits. A finished MP4 with an unwanted expression, a skipped fist bump or distorted hands is a delivered result. Generating again costs credits; there is no free paid-model preview. |
Copy the TV pullout prompt, with both photo roles included
Choose a style below to see its exact 10-second prompt with sound off and no extra note. Image 1 is the room person; image 2 is the TV person. The same prompt builder supplies the generation request.
Create a 10-second vertical 9:16 video of one person pulling another person out of a television into the room. Use exactly two uploaded identity-reference photos: Image 1 defines person A, who stands in the room to the right of the television; Image 2 defines person B, who starts entirely inside the television screen. Keep each person's own face, age, skin tone, hair, body proportions and clothing distinct throughout. Never swap their roles or faces, blend their identities, or add a third person. Stage a softly lit living room with a large television above a low cabinet. Use one steady eye-level wide shot with both people, their faces, arms, hands, legs, the full screen and the cabinet visible. Leave enough space beside the cabinet for both people to stand. Keep the camera fixed for the whole shot; no cuts, zooms, montage, split screen or photo slideshow. Seconds 0–2: person A raises a fist toward person B in the television. Person B reaches one hand through the screen, and their fists touch once at the screen edge. Seconds 2–4: they open their hands and take a secure handhold; person A leans back slightly to help person B forward. Seconds 4–8: person B passes through the screen with subtle violet sparks around its edge, brings each leg over the cabinet in sequence, and steps down onto the floor beside person A. Show the complete exit continuously; neither body may pass through the cabinet or jump instantly from the screen to the room. Seconds 8–10: now together on the floor, the couple smile and share a warm, gentle two-arm embrace. Keep both faces slightly offset and visible at three-quarter angles. No kissing or exaggerated romantic gestures. Maintain exactly two people from start to finish. Preserve believable shoulder, elbow, wrist, hip and knee positions during the fist bump, handhold, exit and hug. Keep hands separate and anatomically plausible, with no fused fingers, extra limbs, intersecting faces or identity changes. No captions, logos or watermarks. Silent video: no audio, no music and no speech.
You do not need to paste this into the tool: it is applied automatically. Keep any optional note compatible with the living room, two people and one continuous shot. The sound switch requests ambient effects, not the song from a particular clip.
Pull out of TV AI trend questions
Can I make the TV pullout video with only one photo of us together?
Crop the shared photo into two separate images, one person per crop, and upload them in role order. Keep both crops at least 300 px on the short side and include shoulders or arms when possible. A tiny face cropped from a group shot is unlikely to give a clear identity reference.
What changes between the couple and friend versions?
Both follow a fist bump, hand grip, step out of the television and final hug. Couple requests a warm embrace; Friends requests a casual, friendly reunion. The photo order, 10-second length and credit price are unchanged. Neither version requests a kiss.
Why does the face change when someone leaves the TV, or the hands stick together?
The model has to reconstruct the face and body as they cross the screen edge, then resolve two touching hands and overlapping arms. Use sharp photos with clear faces, comparable lighting and visible upper bodies. Avoid extra gestures or camera moves. The prompt asks for continuity, but no model setting guarantees perfect faces or fingers.
Can I generate a complete TV hug for free, without a subscription?
A subscription is optional, but generation uses credits. Eligible new-account gifts are up to 5 credits, below the 17-credit minimum. The $5 one-time pack gives 24 credits and covers one Mini 480p video; it does not cover the default 36-credit 720p video. Any completed pack or plan purchase unlocks clean MP4 downloads.
Does this include the original TREND DA TELEVISÃO song?
No fixed song is included. Sound off produces a silent request; sound on requests ambient room and screen effects without music or speech. To match a particular post, add an appropriately licensed track in your publishing app. This page does not claim one song is the definitive original sound.
Is the pull out of TV AI trend Korean version a different effect?
The TV effect here is identified by its action: a hand through the screen, a person stepping into the room, then a hug. A Korean label or soundtrack does not define a separate preset. Stadium broadcast portraits and smart-TV news are different topics; check the actual clip before choosing a tool.
I found the pull out of TV AI trend on Reddit. Can I make it in a browser?
Yes. Use this page on a phone or computer, upload your two photos, choose Couple or Friends and check the displayed credits. No Reddit download, editing-app template or APK is needed. The result is a vertical MP4; your exact faces, motion and background will vary by generation.
Bring the person on screen into the room
Add two photos in role order, choose a reunion style and check the credit price.
Make a TV pullout video