How to Make the TXT Stick With You AI Video
The Stick With You template puts five people into Tomorrow X Together's performance choreography — the tutting, the leg sweeps, the eye-covering move, the formation changes — from five photos.
Upload one photo per person. Photo 1 becomes the main performer (center position). Photos 2 through 5 fill the remaining four spots. The AI rebuilds the choreography with your five people, their outfits, and the original Stick With You audio. Horizontal 16:9, no watermark. If you have used any of our other AI face swap video templates, the flow is the same — just with five upload slots instead of one or two.
The Choreography
"Stick With You" is the title track of TXT's eighth mini album 7TH YEAR: A Moment of Stillness in the Thorns, released in April 2026. The music video hit 10 million views in two days, and the performance version — all five members in black outfits, no set, just choreography and a moving camera — is what this template uses.
The choreo is known for three things:
Tutting. The geometric hand and arm movements that use joints as pivot points. Fingers fold into shapes, arms lock at right angles, and the five members build patterns together that look like one connected machine. MOAs (TXT fans) specifically highlight the hand sequences as some of the most intricate in the group's catalog.
The leg sweeps. Low, fast kicks that move through the formation while the upper body stays locked. Watch the mirrored dance practice on YouTube — the sweep around 0:29 is a standout moment.
The eye-covering move. A gesture where one hand covers an eye while the other arm extends — it showed up in the pre-comeback choreography teaser and became one of the most replicated moves on TikTok.
Five People, Five Photos
The template swaps all five dancers. The AI aims to preserve for each person:
- Their face, hair, and skin tone from the photo
- Their outfit from the photo (replacing the original black styling)
Photo 1 is the main performer — center position, featured in close-ups, leading the formations. Photos 2–5 fill the rest of the group. The mapping is designed to hold through cuts, angles, and formation changes so each person stays in their assigned role.
Who goes where matters. If one of your five has a standout outfit or is the reason you are making the video, put them in Photo 1. That is the slot the camera favors.
Photo Tips for a 5-Person Template
Five photos means five chances for something to go wrong. A few things that help:
- Face and upper body minimum for everyone. The template needs enough visual information to build each person consistently across formation changes.
- Different outfits help. Five people in similar clothing can make it harder for the AI to keep identities distinct. Varied colors and styles give clearer separation.
- Even lighting across all five. Mismatched lighting (one photo in direct sun, another indoors) can create inconsistencies in how the five people look together in the same scene.
- No sunglasses. The choreography includes close-ups and expressive facial acting. Covered eyes limit what the AI can work with.
FAQ
How many photos do I need? Five. One per dancer.
Which photo is the center performer? Photo 1 (labeled "Main performer" in the template). This person gets the center position and the most camera time.
Is it horizontal or vertical? Horizontal — 16:9.
Does the audio play? Yes. The Stick With You track plays over the video.
Do our outfits show up? Yes. Each person's outfit from their photo replaces the original black styling.
Can I use fewer than five photos? The template requires all five slots filled. If you want fewer people, you could use the same photo in multiple slots, but the choreography is built for five distinct dancers. For a 2-person K-pop challenge, check out the Stray Kids This and That template.
Related
Ready to Transform Your Workflow?
Edit photos, color grade entire shoots, and generate AI videos—all in one platform. Just describe what you want in plain English, and Summrs handles the technical work.
Try 10 Photos Free →