AI Services Market
AI Services Market › Fabrique de compétences › Vidéo › Composition Strip
$4.99 · Sans inscriptionv1.0.0– vues

Composition Strip

Faites défiler les cadrages, puis arrêtez-vous sur le bon.

Une bande horizontale de vignettes de composition qui glisse, se parcourt et se stabilise sur l'image gagnante — pour que le public voie comment un plan est composé avant qu'il passe en plein écran.

Aperçu de Composition Strip
Idéal pour
Essayistes vidéo, professeurs de photographie et de cinéma, et designers qui comparent cadrages et recadrages à l'écran
Temps nécessaire
~25 min pour une première bande
Niveau
Accessible aux débutants

Installez-la en 3 étapes

Sans inscription. Sans outils en ligne de commande. Choisissez votre IA, copiez une ligne, collez-la.

Pro — les 511 compétences

Vous préférez le faire vous-même ? Achetez la compétence ($4.99), téléchargez le ZIP depuis votre confirmation d'achat, décompressez-le dans le répertoire indiqué ci-dessus, redémarrez votre session IA — même résultat. Vous l'avez déjà achetée ? Votre assistant IA peut la télécharger pour vous.

Contenu du pack

Un fichier SKILL.md (le guide pratique) + un modèle de départ. Aperçu :

Composition Strip

When to use

  • A video essay or tutorial needs to show several framings of the same shot side by side
  • A design critique must compare compositions — rule of thirds, centring, negative space — without cutting away
  • A photography or film breakdown wants the audience to feel the difference between a near-miss frame and the keeper
  • A portfolio reel needs a fast, elegant way to flip through alternate crops of one image
  • A before-and-after needs to show the original framing, the adjusted framing, and the final frame in one continuous move
  • You have 3–8 frames of the same subject and want the viewer to pick the winner with you

Workflow

  1. Pick the frames and order them (input: your shots or crops → output: strip sequence)
    Choose 3–8 frames of the same subject: the frames you actually considered, not random filler. Order them deliberately — worst-to-best for a reveal, or chronological for a process story. Each frame gets a one-line label ("Centred — static", "Thirds — breathing room") and an aspect ratio; keep all frames the same aspect so the strip reads as one object. The final frame in the strip is the keeper — the one the strip will settle on.
  2. Design the strip layout (input: strip sequence → output: layout)
    The strip is a horizontal row of equal-height thumbnails, edge to edge, with a 4–8px gap between frames. It sits centre-screen at 55–70% of frame height, with a thin baseline or rail beneath it so the strip feels like a physical object sliding on a track. One frame at a time is "active": larger, full-brightness, sharp. The rest sit at 55–65% brightness and slightly desaturated, so the eye knows where to look. Leave generous empty space above and below the strip — the strip needs room to breathe.
  3. Choreograph the slide and scrub (input: layout → output: timing map)
    Total duration 4–7 seconds for the full pass. The strip slides laterally so each frame takes its turn in the active slot: 350–450ms per slide, with a 150–250ms hold on each frame so the viewer can actually read the composition. Easing: ease-in-out cubic (cubic-bezier(0.65, 0, 0.35, 1)) — the strip should glide and settle, never lurch. For a scrub variant, the strip moves continuously under a fixed playhead at 1 frame per 300ms with no holds, then eases to a stop on the keeper.
  4. Set keyframes for the active frame (input: timing map → output: keyframe spec)
    Active-frame keyframes: on entering the active slot, scale 0.94 → 1.0 and brightness 60% → 100% over 250ms with ease-out; on leaving, scale 1.0 → 0.94 and brightness 100% → 60% over 200ms. Add a 2px accent outline that fades in on the active frame (opacity 0 → 1 over 150ms) and a label that slides up 8px and fades in beneath it. The strip itself moves by translateX in exact frame-width-plus-gap steps, so each frame lands dead-centre in the active slot — no half-frame drift.
  5. Land on the keeper (input: keyframe spec → output: finished motion)
    The final slide is the slowest: 500–600ms, ease-out, landing on the keeper with a subtle scale settle (1.0 → 1.03 → 1.0 over 250ms). The keeper's label stays up, the other frames dim a further 10%, and a small "keeper" marker (a dot, a corner tick, or the word itself) fades in. Then the keeper expands to full screen over 500ms (scale up, other frames fading out, strip rail dissolving) — the strip was the selection, the full frame is the payoff. Hold the full frame at least 1 second before cutting.
  6. Check readability at full speed (input: finished motion → output: verified animation)
    Watch at real speed on a phone-sized screen: can you tell the frames apart, read each label during its hold, and spot the keeper the moment it lands? If two frames look identical at strip size, swap one out or zoom the crops apart. Check the strip movement: does it glide in whole-frame steps, or does it stutter? Any stutter means the translateX step math is off — recheck frame width plus gap.

Aperçu du modèle

Composition Strip — Starter Template

Fill in every {placeholder}. All frames: same subject, same aspect ratio, keeper last.

Strip sequence

  • Frame count: {3–8} · Keeper: position {last} — {keeper label}
  • Ordering logic: {worst-to-best / chronological / other}

Layout

  • Strip height: {55–70%} of frame · Gap between frames: {4–8px}
  • Rail beneath strip: {thin baseline, colour}
  • Active frame: brightness {100%}, scale {1.0}, outline {2px accent}
  • Inactive frames: brightness {55–65%}, slightly desaturated
  • Label placement: {beneath active frame, slides up 8px}

Timing map

  • Total duration: {4–7s}
  • Per slide: {350–450ms} · Hold per frame: {150–250ms}

Le pack de compétence lui-même (SKILL.md + modèle) est rédigé en anglais.

Pro — les 511 compétences

Vous aimerez aussi