Quick answer
Starrd's Christmas Dance template turns two photos into the viral TikTok Christmas Dance to DJ Izzy Duzzit's 'Deck the Halls' remix. Photo 1 dances on the left, photo 2 on the right, in a bright bedroom with a little lit Christmas tree, both in matching red-and-white candy-cane-striped Christmas pyjamas (or anything else you ask for in Director's Notes). One continuous take: hand-holding knee kicks and hip sways, matching hand rolls, a few steps toward the camera, then arms up overhead, waving and clapping. Photo 2 sings along from 'Deck the halls with boughs of holly'. Everyone keeps their own face and hair. Works with pets. 14 seconds, vertical 9:16, Seedance 2.5, 480p, 150 credits (about $7.50).
The same dance, any subject — tap one to make it with your photo.
Matching Christmas pyjamas, a little lit tree and DJ Izzy Duzzit's "Deck the Halls" remix: you and a friend hold hands, kick on the beat, walk up to the camera and finish with your arms in the air. Two photos put you (or your pets) in the viral TikTok Christmas Dance. Here it is:
Christmas Dance
Two photos. The viral Deck the Halls duo dance in matching Christmas pyjamas. 14 seconds.
What You Get
- Two dancers. Photo 1 on the left, photo 2 on the right, singing along.
- One continuous take. The same routine as the original, every move on the beat, from a still camera.
- Matching Christmas pyjamas. Red-and-white candy-cane stripes with green holly, or whatever your Director's Notes ask for.
- Your own faces and hair, in a bright Christmas bedroom with a lit tree and stockings.
- People or pets. Dogs and cats dance upright on real paws.
- 14 seconds, vertical 9:16, with the remix, ready for TikTok, Reels and Shorts.
What the Video Actually Is
The clip is from @ayiahsoufi_ on TikTok, posted November 2, 2026 and captioned "should we make a dance?? 👀🎄" (5.7M+ plays, 386K+ likes and 28K+ saves in its first ten days). The sound is DJ Izzy Duzzit's "Deck the Halls" remix, at about 103 BPM. Two friends side by side in a bedroom do a short routine in one locked-off phone shot: hand-holding knee kicks, hand rolls, a walk toward the camera, then arms up overhead, waving and clapping, while one of them sings along to "fa la la la la".
Why it suits AI: it's the Christmas dance everyone's doing this year, and it needs two people, matching pyjamas and a decorated room. Two photos cover all of it, including the version where your dogs do it.
The 14 Seconds, Beat by Beat
| Time | Shot | What happens |
|---|---|---|
| 0:00 | Wide, full figure | Holding hands low, alternating knee kicks and hip sways on the beat; photo 1 flings an arm up at 3.9s |
| 4.1s | Wide, full figure | Matching hand rolls and shimmies in front of the chest; photo 2 starts singing "Deck the halls with boughs of holly" |
| 7.0s | Walking in | Both walk a few steps toward the camera until they're framed from the knees up; photo 1 pulls pouty faces |
| 8.8s | Knees up | "'Tis the season to be jolly", hands and hips moving with the song |
| 10.5s | Medium-wide | Arms up overhead, waving and clapping, knee lifts |
| 13.5s | Medium-wide | Both arms raised high, laughing |
Your Look, Their Moves
- Matching pyjamas by default. Both of you get the same red-and-white candy-cane flannel set with fluffy white socks.
- Or your own outfit. Write it in Director's Notes ("our own clothes", "green elf onesies", "ugly Christmas sweaters") and both of you wear that instead.
- Your own hair. The original dancers' long hair stays out of it; glasses only if your photo has them.
- Pets stay pets. A dog or a cat stands upright on real paws, bare unless your notes dress it.
Photo 2 sings along for most of the video, so put the friend who knows the words second. Photo 1 gets the pouty faces around 8 seconds.
The Fastest Way
Christmas Dance
Two photos, one take: knee kicks, hand rolls and arms up to the Deck the Halls remix. 14 seconds.
Two photos, 150 credits (about $7.50), no prompt. It renders on Seedance 2.5 at 480p, vertical 9:16, 14 seconds, with the remix on top.
Build It Yourself
1. Turn the clip into a depth map. Run the clip through a depth-estimation model (Depth Anything V2 works). Paint out the burned-in caption first: raised hands pass through it, and text in a depth map turns into shapes. The depth map keeps every step and the walk to the camera, and drops the faces, the clothes and the colour, so nobody from the original shows up in your video.
2. Make a storyboard. A vertical 2×3 sheet of six stills of your two subjects in the Christmas bedroom, generated from your photos.
A photorealistic VERTICAL 9:16 storyboard sheet of SIX film stills in a grid of two columns and three rows, separated by thin black lines, with no text. All six stills are frames from one phone video filmed on a camera standing still at chest height in the same cosy bright bedroom at Christmas: white walls, a white ceiling with a ceiling fan, a dark wood floor, a bed with pink bedding on the right and a little lit Christmas tree. The subject from reference image 1 always dances on the LEFT and the subject from reference image 2 always on the RIGHT, both in matching red-and-white candy-cane-striped flannel Christmas pyjamas with a small green holly print and fluffy white socks. Top-left: holding hands low, a knee kick, grinning. Top-right: the left dancer flinging an arm up. Middle-left: matching hand rolls in front of the chest. Middle-right: both walking toward the camera, framed from the knees up. Bottom row: arms up overhead, waving and clapping, then both arms raised high, laughing.
3. Drive it with the depth map, with the song inside. Feed the storyboard, your photos and the depth video (with the song as its audio) to a model that copies motion from a reference video (we recommend Seedance 2.5). Then write what it can't read from a depth map:
- Who is who: photo 1 always on the left, photo 2 always on the right.
- The outfit: one outfit per dancer, written out in full, never "pyjamas or their own clothes".
- The mouths: when photo 2 sings along, and that photo 1 never mouths the words.
Lay the remix back on top at the end: Seedance can drop or change music on its own, so mux the track yourself.
Common Mistakes
Driving with the raw clip. The original dancers' faces and outfits come straight through. Use the depth map.
Leaving the caption in. The burned-in "pov" text sits right where the raised hands go, and it warps the arms in the depth map. Fill it before you run depth.
Putting a choice in the prompt. "Christmas pyjamas unless they want their own clothes" gives the model two outfits, and it blends them. Decide first, then write one.
Both dancers singing. Without explicit windows, both mouths move. Say who sings and that the other one grins.
Letting the camera move. The original is a phone on a stand. Say the camera never moves, or the walk-in turns into a dolly shot.
Related Reading
- 30 AI Christmas Photo Prompts: copy-and-paste prompts for Christmas cards, matching pyjamas and elf portraits, each with an example photo
- How to Make the 'ATM Christmas Wish' AI Video: this season's other Christmas trend, from one photo
- How to Make the Puerto Rican Mami AI Video: another viral duo dance
- How to Make the Wipe Me Down AI Dance Video: a two-friend dance in one take
- How to Make the Dog Coffee Fail AI Video: two pet photos, one disaster
- Viral AI Video Trends 2026: every trend we've turned into a template this year
