Back to MiniMax H3 prompts
Material match cut / First/Last-Frame Image-to-Video

Coffee-to-desert match transition

Match texture, motion, and camera direction before changing scale from one material to another.

Official manual translationTranslated from a MiniMax H3 manual example
Actual H3 output / 15s
Prompt -> result

What this output demonstrates

The camera pushes through coffee foam and cocoa texture until the same granular motion resolves into a wide desert landscape without a hard cut.

Control pattern
Match texture, motion, and camera direction before changing scale from one material to another.
Check in your run
A hard cut, black frame, or visible fracture makes the transformation feel composited.
Official manual example / English teaching translation / source reviewed 2026-08-08Open this example in the H3 manual
Shot structure
0s8s15s
INPUTI2V
ACTION01
CAMERA01
AUDIOON
Working principle

Match texture, motion, and camera direction before changing scale from one material to another.

Watch for

A hard cut, black frame, or visible fracture makes the transformation feel composited.

Adapt this brief

Prompt draft

Begin from Image 1. Push quickly into the milk foam, cocoa particles, and dark coffee texture until the bubbles, powder, and liquid ripples fill the screen and hide every other object. Maintain realistic macro photography with extremely shallow depth of field. Fine powder drifts slowly through backlight while the surface moves between fine sand and fluid. At the instant when the coffee particles, foam contours, and liquid vortex closely match the dune ridges, wind erosion, and airborne sand in Image 2, transform naturally into the desert landscape. Continue the same forward camera movement to reveal the full dunes from Image 2. Do not crack the frame, fade to black, hard cut, add an obvious visual effect, or create a pasted-together look. Keep the transition realistic, quiet, and restrained, as if the same granular material changed scale.

Replace the subject and reference roles first. Change camera, timing, and sound one variable at a time.