Upload your photo
Choose a clear photo of yourself. Keep the face unobstructed and avoid group photos so the video has one clear character to follow.

Your photo. The next trend.
A quiet meal. A curious camera. Curtains closed.
Upload one character photo to replace the person eating behind the window. Genjutsu follows the supplied clip as the camera moves closer, the character notices it, and the curtain closes. The reference video and prompt are ready to use.
Preview the scene and motion before you create your version.
Your generations appear here. You can download a completed video or reuse its photo and resolution.
Upload your character photo above, choose a resolution, and generate your Window Peeper video.
Watch a completed Genjutsu generation made with this reference scene. Upload your own character photo to create your version.
Example 1
A character in a pink shirt eats by the window, notices the camera and closes the curtain.
Choose a clear photo of yourself. Keep the face unobstructed and avoid group photos so the video has one clear character to follow.
Select 480p, 720p or 1080p. The credit total is shown before you generate, with a fixed price for each resolution.
Start the generation and follow its progress in your history. Preview the completed clip, download it, or reuse your photo for another take.
Window Peeper turns the supplied apartment-window clip into a personalized character video. The scene starts outside the building, moves closer to someone eating, and ends when they spot the camera and close the curtain.
Genjutsu uses your photo to guide the character’s face, hair, body and outfit. The reference guides the eating motions, reaction and curtain movement while the instructions preserve the room, window, table and camera framing.
Generate at 480p, 720p or 1080p from the supplied 9:16 reference. The original clip is 14 seconds long, and pricing uses those 14 reference seconds. The total is shown before you generate.
It is an AI Trend Studio template that uses Genjutsu to replace the person in the supplied window-side scene with your character. The original video guides the meal, surprised reaction, camera movement and curtain close.
Use a clear, well-lit image with one visible character. Show the face and the outfit you want to use. Avoid group photos, blurry faces and heavy occlusion.
Yes. Your image guides the replacement character’s face, hairstyle, body, clothing and accessories. The template targets the person eating behind the window throughout the clip.
No. The original video and default prompt are included. Upload one character photo and select a resolution. You can turn on Prompt to edit the instructions.
Browsing the template and watching the reference are free. Generating a video requires credits, with the total shown before submission.
Each generation costs 4200 credits at 480p, 7980 credits at 720p or 18900 credits at 1080p. The fixed price uses 14 reference seconds.
If a generation task is confirmed as failed, the credits charged for that task are returned to your account. Check the status in your trend video history.
Your photo guides the character and the video guides the movement. AI results can vary around hands, food, glass reflections and the closing curtain. Preview the completed video before sharing.
Yes. The workflow includes the supplied reference video’s original soundtrack in the completed video.
You can browse and watch the reference without an account. Sign in with enough credits to generate a video and find the result in your trend history.
SJinn provides Window Peeper as an AI Trend Studio template using Genjutsu and the included reference clip.