From story to shots

How do you turn an idea for a short video into clear shots before filming?

1250 € per session, excl. VAT

In short

This half-day session of 3 hours introduces the visual planning of a short video and provides practice with generative AI. Participants create a six-panel storyboard, specify actions, framing and sounds, then check its clarity, feasibility and the precautions required for the content used.

Why this training?

An idea that is clear when spoken can become confusing on screen: a missing action, a change of viewpoint or text that is too long. A storyboard makes these difficulties visible before you commit a location, props or people. AI can suggest a shot breakdown and rephrase shot instructions, but it can also add unnecessary scenes or scenes that are difficult to film. The exercise teaches participants to select its suggestions rather than confuse visual richness with clarity of message. Continuing case study: The fictitious association « Atelier des Lucioles Bleues » is preparing a 30-second video explaining how to place a book in its book exchange box. The storyboard must show the steps involved, with no identifiable people, using six shots that can be filmed in a single location. Application at work: For a future short video, reuse the six-panel template, replace the fictitious brief with anonymised information and have someone outside the project review the shot breakdown before any filming.

Objectives of this training

  • Formulate the central message of a 30-second video.
  • Distinguish between a wide shot, a medium shot and a close-up.
  • Break down a simple action into six consecutive shots.
  • Specify the action, framing, sound and duration for each panel.
  • Identify visual inconsistencies and issues requiring attention regarding rights.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Communications staff who are new to preparing short videos.
  • Staff who need to explain an action or a service through images.

Prior skills

  • No prior knowledge of AI required.
  • Ability to describe an everyday action in a few sentences; no drawing skills required.

What you will work on

  1. Read an action in images

    Objective: Recognise the useful elements of a panel. Difference between a script, a storyboard and the final video; shot sizes, sequence of actions and separation of image and sound.

  2. Define the story and prompt AI

    Objective: Obtain an initial proposal limited to what is needed. Introduction to generative AI and its limitations; defining the audience, message, constraints and expected format.

  3. Compose the six panels

    Objective: Translate the shot breakdown into usable visual instructions. Schematic sketches, movement arrows, object continuity, shot durations, voices and on-screen text.

  4. Review, safeguard and correct

    Objective: Assess readability and document precautions. Rights to images and music, individuals’ consent, notices appropriate to the distribution context; risks of deepfakes and rejection of deceptive imitations.

What you take away

  • Video brief: audience, message, duration and constraints.
  • Annotated six-panel storyboard.
  • Reusable AI prompt for proposing a shot breakdown.
  • Completed review checklist with corrections.

Method and assessment

  • Short explanations followed by immediate practice using the continuing case study.
  • Individual work followed by work in pairs, with peer review of the outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Arrange access to a tool authorised by your organisation, along with paper or a digital medium for the panels. Prepare a blank template and visual references that can legally be used; exercises remain fictitious or anonymised.

Observable criteria: The six panels make it possible to understand how the book is placed in the box without additional verbal explanation. ; Each panel specifies framing, an action, a duration and a sound cue. ; The durations total 30 seconds and the actions follow one another without unexplained breaks. ; Each shot remains feasible in the location and with the props specified in the brief. ; Visual sources and requirements for rights, consent and notices are explicitly documented..

Frequently asked questions

Do you need to be able to draw?

No. Rectangles, silhouettes and arrows are enough if the framing and action are understandable.

Do you produce a video during the session?

No. The deliverable prepares for filming; it does not include recording or editing.

Does AI have to generate the illustrations?

No. AI is used for the text-based shot breakdown; the panels can be drawn by hand.

Can you use a colleague’s face?

The exercise avoids this. Any subsequent use requires checking consent, applicable rights and the necessary notices, particularly for a synthetic representation.

Last updated: