EN·
← All notes

Midjourney Animation Tutorial (Edition 1)

Think Like a Director: Writing Cinematic Prompts for Motion

AI × Animation · · Kuo

ShareLinkedIn

Midjourney can animate now — but every motion starts with a single frame.

If you want your animations to feel cinematic, emotional, and intentional, you need to direct the image first. Think like a filmmaker: frame the shot, set the mood, and guide the viewer’s eye.

This guide shares a two-step method and a cinematic prompt framework I’ve been developing through my own experiments. It’s still evolving — but even in its early form, it has helped me create Midjourney animations with more control, clarity, and emotion. If you’re just starting out, it will help you too.

Step 1: Create a Cinematic Image

Before animation, you need a strong image — one that feels like a scene, not just a subject.

Think in scenes, not just subjects.

Most beginners describe what they want to see. Directors describe what they want the audience to feel.

Instead of:

“A girl in a forest”

try:

A girl stands alone in a foggy forest at dawn.
Camera: wide shot from behind.
Lighting: soft morning light filtering through trees.
Atmosphere: mist swirling, quiet tension.

This isn’t just a prompt — it’s a scene. And that’s the shift.

Use the Cinematic Prompt Formula

Here’s a flexible formula I use to build prompts that animate well:

[Subject + Action or Pose]
Camera: [Framing + Angle + Lens]
Environment: [Setting + Spatial Layout]
Lighting: [Source + Direction + Mood]
Atmosphere: [Motion Cues + Texture + Style]

You don’t need to write every label — just think through each layer. This helps you control what the viewer sees, feels, and remembers.

Example: The Samurai’s Gaze

A lone samurai kneels in the snow, head bowed, shot from behind at a low angle.
A lone samurai kneels in the snow, head bowed. Camera: back, low angle. Environment: snow-covered forest clearing, sword planted in the ground. Lighting: overcast sky, soft diffused light. Atmosphere: snowflakes falling, breath visible, cinematic tone, desaturated palette. --ar 16:9 --stylize 150

Each element locks in visual storytelling, guiding the viewer’s eye and emotion.


Step 2: From Still to Motion — The Two-Step Workflow

Once your image is strong, you can evolve it into a short sequence. Think of it like storyboarding: each beat is a frame, and together they form a moment.

Two-step method:

  1. Direct a strong cinematic image. Define subject, camera, environment, lighting, and atmosphere using the formula. Lock in composition, mood, and visual identity.
  2. Adapt the image into a motion-aware prompt. Break the scene into 2–3 visual beats. Add subtle camera movement or transitions. Introduce motion cues (wind, breath, or facial-expression shifts). Use --video and --bs in Midjourney to animate.

Example 1: The Kite Runner

Image Prompt:

Watercolor illustration of a boy chasing his kite across a windy beach.
Wide establishing shot of the ocean and the boy running.
Bright, hopeful atmosphere.
--ar 3:2 --quality 2 --stylize 150
Illustration of a boy running along a windy beach, chasing a kite, wide shot showing the ocean.
The first frame anchors our scene, mood, and composition.

Animation Prompt:

Same scene, but now with shot transitions:
wide establishing shot of the ocean, cut to medium shot of the boy running, close-up of his hand gripping the string.
--ar 3:2 --bs 1 --video 1
Animated sequence of a boy running on a windy beach, following his kite, with close-ups and medium shots.
2–3 beats make this simple scene feel cinematic.

Example 2: The Samurai’s Gaze

Image Prompt:

A lone samurai kneels in the snow, head bowed.
Camera: side view, low angle.
Environment: snow-covered forest clearing.
Lighting: overcast sky, soft diffused light.
Atmosphere: snowflakes falling, breath visible, cinematic tone.
--ar 16:9 --stylize 150
A kneeling samurai in a snow-covered forest, soft snowfall, overcast lighting.
Each detail establishes space, mood, and motion potential.

Animation Variant 1:

The samurai lifts his head; zoom-in on the upper body, snow blowing gently.
The samurai lifts his head and turns slightly to face the camera. Camera: zoom-in on upper body.

Animation Variant 2:

The samurai lifts his head; zoom-in on the face, snow blowing gently.
Same motion, but with a facial-expression shift. Camera: zoom-in on face. Atmosphere: wind picks up slightly, expression changes to calm determination.

Comparison

Side-by-side comparison of the samurai animation variants.
Tip: for a 5-second clip, 2–3 distinct beats are ideal. If motion is subtle, use expression or atmosphere to convey progression.

Wrapping Up: Think Like a Director

With these steps, you’ve built a foundation:

Every word in your prompt is a decision: a camera angle, a shaft of light, a subtle motion. This is how directors think. And now, so do you.

This is Edition 1 of a series on AI video prompting — adapted from my upcoming handbook. Later editions go deeper into constructing scenes, camera movement, and motion techniques.

ShareLinkedIn
← All notes