Quick answer
Starrd's Iconic By Mistake template puts you and two friends in the viral 'Iconic By Mistake' dance trend. Upload three photos: photo 1 dances on the left, photo 2 is the lead in the centre, photo 3 dances on the right. All three hit the routine in unison on a studio stage with a wooden floor and black curtains: hands on hips, an OK sign by the eye, a big jump with the arms thrown up, hands framing the face, a swaying line with linked arms, sharp angled poses on the 'I-I-I am iconic by mistake' hook, then a walk to the camera and a proud final huddle. It's one continuous take, every move is copied from the original clip as a depth map so it lands on the same beat, and everyone keeps their own face, hair and outfit. Pets work too, dancing upright on their hind legs with their hats, bows and glasses on. 22 seconds, vertical 9:16, Seedance 2.5, 480p, 250 credits (about $12.50).
The same dance, any subject — tap one to make it with your photo.
Three of you, one stage, one routine in perfect unison: the OK sign, the jump, the linked arms and the hook poses. Three photos put you and two friends (or three kittens) in the viral Iconic By Mistake dance. Here it is:
Iconic By Mistake
Three photos. The full Iconic By Mistake routine in unison on a studio stage. 22 seconds.
What You Get
- Three of you, one take. No cuts: the whole routine plays in a single shot that opens close and slowly pulls back to a wide.
- A lead in the centre. Photo 2 stands a step in front at the start, like the centre of a pop group.
- The full routine. OK sign, jump, face frame, linked-arm sway, both hooks and the final huddle, on the same beats as the original.
- Your own look. Your faces, your hair and the clothes from your photos, on the clip's studio stage.
- People or pets. Cats and dogs dance upright on their hind legs, accessories on.
- 22 seconds, vertical 9:16, with the song, ready for TikTok, Reels and Shorts.
What the Video Actually Is
The clip is a repost by Billboard Brasil (@billboardbrasil) on TikTok from June 16, 2026, tagged #katseye and #iconicbymistake (626K+ plays, 80K+ likes and 3.3K saves so far). Three dancers in black tube tops and baggy low-rise jeans stand in a line on a studio stage with a wooden floor and black curtains, and run the routine to the "I-I-I am iconic by mistake" hook in perfect unison. It's one unbroken take: the camera starts close on the three of them and slowly pulls back to a wide shot. The track runs at about 150 BPM, so each bar of four beats lasts 1.6 seconds, and the moves change on the bar.
Why it suits AI: a trio routine needs three people who know the moves, the same timing and a stage. Most friend groups have none of those. Three photos cover all of it.
The 22 Seconds, Move by Move
| Time | What happens |
|---|---|
| 0:00 | All three stand with their hands on their hips, chins up, cool and unbothered |
| 1.0s | One hand comes up into an OK sign beside the eye, with a knowing head tilt |
| 2.4s | The big jump: both arms thrown overhead, one knee kicked high |
| 3.5s | They land wide with both hands framing the face; the camera starts to pull back |
| 5.0s | Arms swing out to the sides, then a deep bent-knee stance with the hands at the chest |
| 7.5s | They step into a tight line, arms linked at shoulder height, swaying together |
| 10.0s | Hands on hips, then palms pressed together at the chest, bouncing on the beat |
| 12.5s | First hook: they turn sideways and hit sharp angled arm poses |
| 15.0s | Second hook: one arm flexed beside the head, hips popping from pose to pose |
| 17.5s | They walk forward together towards the camera, flipping their hair |
| 20.0s | A shoulder-to-shoulder huddle, hands on hips, held to the last frame |
Your Look, Their Moves
- Your own outfits. A blazer, a fedora, a T-shirt: whatever is in each photo dances. Nobody gets the original's tube tops or baggy jeans.
- Seen head to toe. All three dancers are shown full length, so if a photo stops at the waist, the rest of the outfit is filled in to match its style.
- Your own hair. The original dancers' long hair stays out of it, so your hair keeps its own cut and length.
- Pet accessories stay on. A beret, a bow or a pair of glasses in the photo stays on through the jump and every pose.
Clear, well-lit photos where the faces are easy to see work best, and full-length ones keep your shoes and trousers exactly yours. Put your boldest friend second: photo 2 is the lead in the centre.
The Fastest Way
Iconic By Mistake
Three photos, one take: the full routine in unison on a studio stage. 22 seconds.
Three photos, 250 credits (about $12.50), no prompt. It renders on Seedance 2.5 at 480p, vertical 9:16, 22 seconds, with the song 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). The depth map keeps all three dancers' moves, the camera's slow pull-back and the shape of the stage, and drops the faces, the clothes and the colour, so nobody from the original shows up in your video.
2. Make one character sheet. A vertical sheet with the empty stage on top and a four-view turnaround (front, three-quarter, side, back) of each of your three subjects below it, generated from your photos. It tells the video model exactly what each dancer looks like from every angle, including the back during the turns.
A photorealistic VERTICAL 9:16 CHARACTER REFERENCE SHEET laid out as four horizontal bands separated by thin black lines, with no text. Top band: an empty dance studio stage with a warm wooden floor, a low slate-blue wall band and tall black curtains. Below it, one band per subject: a 360-degree turnaround of the subject from reference image 1, then reference image 2, then reference image 3, full figure from head to feet, front, three-quarter, side and back, standing relaxed with the mouth closed, each in exactly the outfit from their own photo, with the stage softly out of focus behind them.
3. Drive it with the depth map. Feed the sheet, your photos and the depth video 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: which figure each photo plays, and that they never swap places.
- The moves, with times: a timed list of the routine, so each pose lands on its bar.
- The faces: smug on the OK sign, big energy on the jump, confident in the huddle, because the faces get small in the wide shot and carry no expression through a depth map.
Lay the song 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, hair and outfits come straight through. Use the depth map.
Leaving the original clothes in the prompt's blind spot. The depth map still carries the outline of the baggy jeans and long hair. Say plainly that those shapes aren't anyone's clothes or hair, or tube tops start appearing.
Asking for lip sync. The faces are tiny once the camera pulls back, so the depth map gives the model no mouths to copy. Say nobody sings, or the mouths flap at random.
Pets on all fours. Without an explicit line, the model drops animals back onto four legs. Say they stand upright on their hind legs from the first frame, with real paws, never hands.
Swapping places. Three subjects on a line drift into each other's spots during the linked-arm sway. Pin each photo to left, centre or right for the whole take.
Related Reading
- How to Make the KATSEYE 'Animal' Dance AI Video: the other trio girl-group routine, danced where your photo was taken
- How to Make the 'What's Up In A Minute' AI Dance Video: another three-dancer routine, in four shots
- How to Make the Wipe Me Down AI Dance Video: a two-friend dance in one take
- How to Make the Lonyo AI Dance Video: lead a whole crew, filmed from above
- How to Make an AI Cat Video: more ways to put your cat in a video
- Viral AI Video Trends 2026: every trend we've turned into a template this year
