Raindance AI trend: how to make the pier duet from two photos

To do the Raindance AI trend, you need two clear solo photos: one of each adult who has agreed to appear in the clip. Framecomet's Raindance effect places the first person on the left and the second on the right at the end of a weathered wooden pier at golden hour, then animates them swaying and softly singing toward each other. The result is a vertical 9:16 video of about 10 seconds, and it is silent. You do not film anything, install an app, hunt for a CapCut template or write a prompt. Check the status on the Raindance effect page before you prepare your photos.

What the Raindance trend looks like

The trend recreates a romantic duet moment. Two people face each other at the end of a pier, the sun sits low over a calm sea and a light breeze moves their hair and clothes. The camera drifts slowly toward them, and a warm haze with soft film grain makes the shot feel like a still from a film rather than a phone video.

Framecomet rebuilds that setting from your photos. It first creates a single film-style frame of the two of you in coordinated summer outfits, then animates that frame. No original footage, performers or recordings are used, and nothing is face-swapped onto an existing video. The singing is visual only, because the clip is silent.

What you need before you start

• Two separate photos with one person in each. Person 1 appears on the left and Person 2 on the right.
• Two adults (18+) who have both agreed to be included.
• Photos you took yourself or have permission to use. Celebrities, public figures and minors are not allowed.
• A clear, front-facing view of each face, without sunglasses, masks or heavy filters.
• An account and enough credits for one attempt. The cost is shown before you confirm.

How to make the Raindance AI video, step by step

1. Open the Raindance effect page and check its status.
2. Add a photo for Person 1 (left) and Person 2 (right). You can pick them on your phone or computer before signing in; they stay in that browser for up to 24 hours and upload only after you sign in.
3. Sign in, confirm that everyone pictured is an adult who agreed to be included, and check that each photo sits in the right role.
4. Review the credit cost and clip length shown in the confirmation step, then confirm. Credits are charged only when you confirm and the task is created.
5. Wait while the scene is generated. Your photos, the first frame and the finished clip are each screened by automated safety checks, and some results wait for a human review, usually within a day.
6. Download the clip from your history. Results are kept for 30 days, so save your own copy.

Choosing two photos that work together

Both people share one frame, so the two photos should be roughly comparable. If one photo is sharp and the other is dark and blurry, the second person has much less detail to work from, and that difference tends to show in the result. These habits help:

• Use recent solo photos with the whole face visible and the eyes open.
• Prefer soft daylight or even indoor light over harsh flash or strong colored light.
• Head-and-shoulders or half-body framing works well; tiny faces in wide shots carry little detail.
• Do not use one group photo for both people. Each role needs its own photo so the two identities stay distinct.
• Agree on positions before uploading. The first photo is always the person on the left.

For the full list, including file formats and consent, see best photos for AI trend videos.

What to expect from the result

You get one vertical 9:16 video of about 10 seconds; the exact length appears next to the cost before you confirm. Both people sway gently and appear to sing softly to each other while the camera pushes in. There is no song, no voice and no lip-sync, so mouth movement will not match any lyrics. The clip is silent.

Your photos guide how each person looks, but faces, hands and the way the two of you interact can vary from one attempt to the next, and exact likeness is not guaranteed. Watch the whole clip and check both people before you share it. Small drifts in a face or a hand are a known limit of generated video, so judge the result on the full playback rather than the first frame.

Adding sound to a silent clip

The download is silent by design. If you want sound on TikTok or Reels, add your own licensed audio in your editor, or use a sound the platform offers you under its own terms. Because there is no lip-sync, the clip works best with audio that does not depend on exact mouth movement.

Before posting, show the clip to the other person and make sure they are happy for it to go public. Label it as AI-generated wherever the platform asks, and do not present it as real footage of the two of you.

Common questions

Can we use one photo of the two of us together? Use two solo photos instead. Each role takes exactly one person, which helps keep the two identities from blending.

Does it have to be a couple? No. Any two consenting adults can take part, such as friends or a creator duo. The mood is romantic, so make sure both people are comfortable with that first.

Can I change the location or the outfits? Not in this effect. The golden-hour pier and the coordinated summer outfits are part of the scene.

Is any original audio included? No. The clip is silent, and no original footage, voice or recording is used.

Ready to try it? Check the status on the Raindance effect page, compare it with the two-person Hotel Lobby effect, and read how credits work before you buy a pack.