Section 7: Animation and Image to Video · Lesson 2 of 5 (32 of 45) · 60 minutes · Evidence: Motion prompt tests
Imagine you are directing a very literal actor who cannot ask questions. If you say “do everything at once, dramatically,” you will get chaos. If you say “turn your head slowly toward the window and stop,” you will get what you asked for. A video generator behaves like that actor, so today the skill is saying less, and saying it more precisely.
Production journal: write your answer below.
What must a filmmaker understand before they can write concise prompts that separate subject action from environmental movement?
In earlier sections your prompts described a picture. A motion prompt describes what happens to a picture over a few seconds. Because the image already shows who and where, the prompt does not need to repeat it. It needs to say what changes, how much and how fast.
A subject action is what the character or object does: lifts a lantern, turns, smiles. Environmental motion is what the world does: fog drifting, rain falling, leaves stirring. Keep them in separate phrases, with the subject action first, because the main action is what the tool should protect. Mixed together, “Mara walks in wild wind and flying birds and crashing waves” gives the tool no priority and the result usually shows it.
A constraint limits the motion so it stays controllable. Useful constraints include “face stays the same,” “background stays still,” “no new people appear,” and “camera does not move.” Use two or three, not ten. Each one trims the chaos, but a long list of negatives can confuse as much as it helps.
A useful habit: say the prompt out loud in one breath before you type it. It will tell you quickly whether it is too long. Starting image: Mara at the top of the lighthouse stairs, holding a lamp. Overloaded prompt: “Mara walks down the stairs while wind blows her coat and scarf, the lamp swings, waves crash below, birds fly past, camera spins.” Six actions, no priority. Revised: “Mara slowly lifts the lamp to eye level. Her scarf moves a little in the wind. Constraints: face and glasses stay the same, camera holds still.” One subject action, one supporting motion, two constraints. Small, controlled motion beats chaos, because the viewer reads it and the tool can deliver it.
People who make short animated clips from stills learn quickly that the first prompt rarely works and that the fix is usually fewer words, not more. Study the loop and look for who is steering at each step.
The creator starts with an approved still and a plain-language description of the action. They choose the length of the clip, often only a few seconds, and then generate. The tool returns a clip, and from here the creator’s attention is on what went wrong, not on whether it is pretty: did the face change, did the hand bend, did something appear that was not asked for? They then adjust one thing in the prompt and run it again.
Pay attention to two habits. First, careful creators change one variable at a time, so they learn what each phrase does. Second, they keep a record of each attempt, including the failed ones, because a failure often shows exactly which word caused the problem.
As you study, record each decision point in your production journal.
| Decision point | What the filmmaker does | What the tool automates | What still needs human review |
|---|---|---|---|
| Pick the image and action | Chooses one approved still and one main action | Nothing; the tool waits for instructions | Whether the image gives the action a clear place to happen |
| Write and run the prompt | Types a short prompt with action, amount, speed and limits | Reads the prompt and generates a short clip | Whether the prompt says one thing clearly or asks for too much |
| Inspect the result | Plays the clip at normal speed and then slowly, checking faces and hands | Shows the finished clip | Warping, extra objects, changed identity or meaning |
| Adjust one thing and retry | Edits a single phrase and runs it again | Produces a new clip | Whether the change helped, which is only knowable by comparing both clips |
Studio task: Revise overloaded prompts into one controlled action plus supporting motion.
Your teacher gives you three overloaded motion prompts, such as one that asks for a walk, a storm, flying birds, a swinging lamp and a spinning camera in the same clip. Work with a partner.
If your school provides a generation tool, you may test one of your rewrites on an approved image. If not, you can still do every step above on paper and test with a simple zoom in your editor.
Coaching tips. Read your rewrite aloud as if you were cueing an actor. If you run out of breath, it is too long. If two verbs are joined by “and,” ask which one the story needs and move the other to a second, later clip.
Test number: ____ Source image (file name and where it came from): ______ Main subject action: ______ Supporting environmental motion (one at most): ______ How much / how fast: ______ Constraints (two or three): ______ Full prompt as typed: ______ Result in one sentence: ______ Keep or retry, and the one thing I will change next: ______
Studio task: Test three shot prompts and log the results.
Choose three shots from your motion plan in 7.1 that you marked Move. If you have access to a school-approved generation tool, test all three. If not, write the prompts, then simulate the motion with a slow zoom or small shift on the still in your editor, and treat that as your result.
Checklist before you submit:
Underline the verb that controls the main action.
Deliverable: Motion prompt tests.
Submit the motion prompt tests. The file must use the course naming convention and include enough process evidence to show the student's decisions and revisions.
File name (course naming convention): AIF_S07-L02_motion-prompt-tests_v01_YOURINITIALS
| Criterion | Advanced | Proficient | Developing | Beginning |
|---|---|---|---|---|
| The work addresses the stated story or production purpose. | Each prompt names one story-driven main action and the log explains how the motion supports the scene. | Prompts target a clear story purpose in each shot. | Purpose is stated for some shots but not tied to the scene. | Prompts are unrelated to the film’s story. |
| The student applies the required method accurately and produces usable capstone evidence. | Prompts separate subject action, environmental motion, amount, speed and constraints, and are reusable by a teammate. | Prompts have a main action and at least one constraint, and are logged in full. | Prompts mix several actions or omit constraints. | Prompts are not logged or cannot be reused. |
| The student identifies at least one quality issue and records a meaningful revision decision. | Changes one variable between tests and shows with evidence whether the revision helped. | Records a revision based on a visible flaw. | Notes a flaw but changes several things at once. | No revision recorded. |
| The work complies with consent rights attribution privacy disclosure and school policy. | Records the source of every image, avoids protected characters and real people, and documents a no-cost route. | Records sources and avoids protected material. | Some sources are missing. | Uses protected characters or a real person’s likeness without permission. |
Requirement: Do not request copyrighted characters or nonconsensual real-person animation.
A motion prompt can turn a still into a convincing moving image, so be careful about whose image you animate. Use only your own characters, licensed material, or a person who has given written consent. Do not ask a tool to make a real person do or say something they did not, and do not request movements from a character that belongs to someone else. Log each test with its source image. If your school gives you limited access to a generation tool, spend it on your three most important shots, and use the editor route for the rest.