Your photos. Three dancers. One Animal routine.You and your friends, one Animal dance trio. Upload your photos to recast the three dancers—or make three versions of yourself.
The You Move Like an Animal AI format recasts a three-person routine associated with KATSEYE's Animal. Your photos define the people; a reference clip guides the dance.
The dark studio and pale wooden floor come from KATSEYE's official three-person short. This template uses 13–28 seconds: the center/right exchange, synchronized gestures and steps, then a forward approach. Your photos recast the dancers while the clip guides their motion.
Watch the official KATSEYE three-person shortThe template keeps the reference setting. It does not move the trio into the background of your first photo.
The song, the three-person photo format, and this template are related—but they are not the same thing.
KATSEYE released Animal with an official music video. The phrase You Move Like an Animal is another name people use to find the song and related dance edits.
KATSEYE Animal · Official releaseAI template guides describe a trio format made from one to three photos. A single person can fill the whole lineup, or different people can take the three roles. Versions differ in length, setting and model.
Read a guide to the trio formatThis template selects 15 seconds from the official studio short. Match photos to the selected clip's opening positions—not the dancers' positions at the end—then review each identity through the exchange.
Watch the official KATSEYE three-person shortChoose your lineup, check the preview, then make your trio.
Assign one clear photo to each starting role: Center, Left and Right. Use one photo for three versions of yourself, or add friends to the lineup.
Watch the 15-second reference and choose 480P or 720P. The generation button shows the total cost for all three dancers before you submit.
Check each face and outfit before and after the center/right exchange, as well as turns and overlaps. Download your video or open it again from History.
Give each role its own clear reference. A full-body photo helps define clothing and proportions that a tight selfie cannot show.
Set the lineup without writing a custom dance prompt.
Assign photos by their starting positions, viewed from the camera. These labels bind a person to a dancer, not permanently to one side of the screen.
Make three versions of yourself, or add separate photos for a friend lineup. The role labels show which dancers reuse the same photo.
The prompt requests each person's face, hair, body proportions and visible clothing from their own photo. Unseen details still need to be inferred.
The video supplies the movement and setting. You can inspect the routine before generating instead of relying on a song title alone.
The total covers one 15-second video with all three roles, whether you use one photo or three.
480P uses 225 credits; 720P uses 375 credits. Check the current total on the generation button before submitting.
Your cash cost depends on your plan or credit pack. A new generation is a separate attempt and spends additional credits.
Generation uses Wan3 Prime through fal. Technical failures return generation credits automatically; visual imperfections in a completed video are not a technical failure. View the model documentation
The shared song name does not mean every tool makes the same video. Check the actual motion, inputs and output before choosing.
Explore more photo-based routines with the same account.
It is a photo-based dance remake format associated with KATSEYE's Animal. This page recasts the three dancers in the displayed reference using your photos. It is not a tool for creating a new song or a general animal-animation effect.
Yes. With one photo, the same person fills all three roles. With two, photo 1 fills Center and Right while photo 2 fills Left. With three, photo 1 is Center, photo 2 is Left and photo 3 is Right. Check the lineup before submitting.
No. The labels describe the opening lineup of the selected 15-second clip, from the viewer's perspective. The center and right dancers exchange positions; each assigned identity should stay with its dancer. Crossing and occlusion can still cause errors, so review the full output.
The prompt requests each person's face, hair, proportions and visible clothing from their own photo. The setting comes from the reference video, not your photo's background. Clothing details and rear views can change, especially when the input is only a portrait.
No. This is a 15-second excerpt, covering 13–28s of the official studio short. It includes the center/right exchange and ends on the forward approach, before the full short's final pose. Choose 480P for 225 credits or 720P for 375 credits.
The reference guides motion and timing, but AI generation is not frame-perfect copying. Fast head turns, overlapping bodies and hand movements can change faces, clothing or anatomy. The preview helps you judge the intended routine; review your own full result too.
The finished video uses the selected reference's audio; display previews may be silent. This is not voice cloning or AI songwriting. Generating a video does not grant commercial music or footage rights.
Technical failures return generation credits automatically. A completed video with imperfect likeness or movement is not a technical failure. Generating another version uses additional credits.
This version is for people and human-shaped characters in the fixed trio reference. It does not offer a pet mode, a new background or custom choreography. Use AI Character Swap for a workflow with your own reference video and instructions.
No. This is an independent AI character-replacement tool, not an official KATSEYE product or endorsement. Review your result and disclose realistic AI edits when sharing.
