Film-scene motion transfer
End of Watch AI Video Generator
Recreate the End of Watch locker scene with two photos. Your images guide both the foreground character and the friend behind them while the source clip guides the camera, gestures and timing.
Create your video- Use two photos to replace both on-screen partners
- Follow the original camera and gesture sequence
- Preview the source clip before creating
The video shown is the original source clip, not a generated result.
Missing 2 referenced images — upload them for better results.
Examples
Original scene preview
This is the Instagram source clip used to show the scene and timing. It is not an AI-generated result.
Original locker-room clip
The source scene shows two partners in front of blue lockers and a sequence of close-camera prop gestures.
How it works
How it works
Review the source scene, then upload one clear photo for each of the two roles.
Upload two character photos
Upload separate clear photos for the foreground person and the friend behind them. Each photo should show one unobstructed face and upper body.
Review the scene
Watch the original reference clip and check the selected model, duration, resolution and credit cost.
Generate and inspect
Generate the video, then check identity consistency, hands, prop transitions and the partner's continuity before sharing.
About the trend
About the End of Watch scene
This template is based on a locker-room clip from the 2012 film End of Watch. Add two character photos to place yourself and a friend in the two on-screen roles. The Instagram edit jokes about two friends whose next business might not work out. Original post: https://www.instagram.com/p/DE1LlNjN00R/. The uploaded clip is the original scene reference, not an AI-generated result.
Frequently asked questions
Frequently asked questions
What to expect from this reference-video template.
Which character does each photo replace?
Image 1 guides the person nearest the camera. Image 2 guides the partner who appears behind them. Upload a separate photo of one person for each role.
Is the preview an AI-generated result?
No. The preview is the original Instagram source clip, provided to show the scene and its timing.
Can I upload one group photo instead of two photos?
This preset expects two separate character photos, one for each on-screen role. A group photo does not reliably assign the two identities.
Will the output match every frame exactly?
The prompt asks the model to follow the reference closely, but generated motion, faces, hands, objects and audio may vary. Review the result before using it.
