Assign authority.
Then stage the shot.

[REFERENCE USE]
Image1–Image6 define only the available characters, objects, and locations. Do not copy their framing.
Video1 defines shot rhythm, transition language, and music only. Do not copy its subjects.

[SCENE]
Create one coherent cinematic sequence using the six image assets. Preserve each asset’s defining appearance and keep subject count stable.

[EDITING]
Match Video1’s pace and transition grammar without recreating its story. Keep geography legible and every transition motivated.

Image1 defines the coffee surface: milk foam, cocoa dust, bubbles, and dark liquid.
Image2 defines the destination desert and its wind-carved dunes.

One continuous photoreal macro shot. Push rapidly into Image1 until particles, foam contours, and ripples fill frame. At the moment those forms align with the dune ridges and airborne sand in Image2, transform seamlessly into the desert and continue pushing forward to reveal the full landscape.

Extremely shallow depth of field at the start; quiet restrained sound. No hard cut, black frame, tear, or visible compositing seam.

Image1 defines the two characters’ identities and wardrobe. Ignore its background and pose.
Video1 defines body motion, expressions, interaction timing, and camera framing only. Do not use its people.

At the sink on frame right, the man hands a clean plate to the woman on frame left. He turns and flicks dish-soap foam toward her with his right hand. She recoils, then immediately retaliates. They laugh, dodge, and throw foam back and forth.

Keep both identities stable, preserve left/right geography, and make foam, wet surfaces, shadows, and reactions physically believable.

Video1 is the sole reference for motion paths, timing, and the locked-off wide camera. Replace its three suited men with exactly three photoreal capybaras.

Preserve the action order: all three drop to the floor; the left capybara jumps to center; the center capybara rolls to far left; the new center capybara rolls to far right; the right capybara jumps to center; finally the center capybara jumps onto the other two to form a pyramid.

Keep the camera fixed. Integrate fur, weight, floor contact, lighting, and shadows naturally. No extra animals or position swaps.

Video1 defines the character, scene, camera, and visible performance.
Audio1 defines only the character’s voice timbre, cadence, and emotional delivery.

The character says exactly: “Follow the wind, live free. Leave worries behind and enjoy the moment.”

Match Audio1’s voice while preserving clear lip sync, the original character identity, and native stereo ambience. No subtitles, extra speech, or voice drift.

[SOURCE MASTER]
Video1 defines the subject, performance, foreground timing, camera, and duration.
Video2 defines only the target fairy-tale environment, atmosphere, and lighting style.

[EDIT]
Remove the green background in Video1 and replace it with the environment from Video2. Make background motion and parallax respond correctly to the subject. Relight the subject to match the new scene.

[PRESERVE]
Keep identity, hair and fabric edges, gesture timing, scale, camera, and duration unchanged. No green spill or altered performance.