AI Services Market
AI Services Market › Fabrique de compétences › Vidéo › Center and Corner Tags
$4.99 · Sans inscriptionv1.0.0– vues

Center and Corner Tags

Un titre en plein centre, quatre faits dans les coins — qui arrivent en décalage délibéré.

Apprend à votre IA à composer des cartes de titre où une étiquette centrale arrive d'abord et quatre étiquettes de coin arrivent en décalage diagonal, avec géométrie, timing, hiérarchie et règles de sortie exacts.

Aperçu de Center and Corner Tags
Idéal pour
Créateurs vidéo, pédagogues et présentateurs qui ont besoin de cartes de titre lisibles en un coup d'œil
Temps nécessaire
~20 min pour un premier brouillon
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 :

Center and Corner Tags

When to use

  • A video title card where one main topic needs four supporting keywords arranged around it
  • A product feature callout where the hero claim sits center and four specs anchor the corners
  • A course or documentary section opener where the chapter name is central and four sub-topics preview at the edges
  • A livestream or presentation bumper that must be readable in the first two seconds, even on a phone
  • A social clip intro where the hook is central and four tags tease what the video will cover
  • A portfolio reel divider where the project name is central and role, year, client, and medium sit in the corners

Workflow

  1. Lock the five labels (input: your topic and four supporting points → output: five final label strings)
    Write the center label first: one short phrase, 2–5 words, the single thing the viewer must remember. Then write exactly four corner labels, each 1–3 words — no sentences, no filler words. Cut any corner label that repeats the center label; each corner must add one new fact. If you cannot name four distinct facts, use three corners and leave the fourth empty rather than inventing a weak label.
  2. Set the frame geometry (input: five labels → output: layout grid with safe margins)
    Place the center label at the exact optical center of the frame (50% / 50%, nudged up by ~2% of frame height if the type has a heavy descender line). Place corner labels inset from the edges by a single consistent margin — 6% of frame width on the sides, 8% of frame height top and bottom — so all four sit on the same invisible inset rectangle. Corner labels align outward to their corner: top-left aligns left, top-right aligns right, bottom-left aligns left, bottom-right aligns right. Cap the center label at 70% of frame width and each corner label at 22% so nothing collides on a vertical phone crop.
  3. Choreograph the stagger (input: layout grid → output: timed entrance sequence)
    Animate in this order: center first, then corners in a deliberate sequence — top-left, bottom-right, top-right, bottom-left (a diagonal zigzag that keeps the eye moving across the frame rather than circling it). Center: 0.00–0.45s, scale 0.92 → 1.00 with a soft overshoot, opacity 0 → 100. Each corner: 0.30s duration, starting 0.12s after the previous element began, so entrances overlap slightly but never land simultaneously. Corner motion is a 12px slide from its own corner direction plus opacity 0 → 100 — no rotation, no bounce. Total entrance: ~1.1s. Then hold all five labels static for at least 1.2s before anything else moves.
  4. Style the hierarchy (input: timed sequence → output: type and color spec)
    The center label is the only bold, large element: roughly 3× the corner point size, high-contrast fill (near-white on dark, near-black on light). Corner labels are one weight lighter, 60–70% of the center label's visual weight, set in small caps or with a thin rule above or below them so they read as tags, not headlines. Keep all five labels in one typeface family; vary weight and size, never the family. If the background is footage, add a soft dark vignette or a subtle scrim behind the labels — never a heavy solid box that hides the footage.
  5. Build the exit (input: finished entrance and hold → output: complete animation with exit)
    Exits run in reverse stagger: corners leave first (bottom-left, top-right, bottom-right, top-left), each 0.22s, fading and sliding 8px back toward its corner, 0.06s apart. The center label exits last: 0.35s, scale 1.00 → 0.96, opacity 100 → 0. Never cut straight from the hold to the next scene — the exit gives the viewer a clean visual breath and prevents the hard jump that makes tag cards feel cheap.
  6. Adapt and export (input: complete animation → output: platform-ready renders)
    Re-check the layout on the narrowest target aspect ratio first (9:16 vertical): corner labels must not collide with platform UI overlays, so keep them inside the central 85% safe zone for vertical. If a corner label is cut off, shorten the label — do not shrink the margin. Export at the source frame rate (24/25/30fps, never converted), ProRes or high-bitrate H.264 for masters, and a captioned version if the labels carry information a muted viewer needs.

Aperçu du modèle

Center and Corner Tags — Starter Template

Fill in every {placeholder}, then hand this spec to your video tool or motion designer.

Project

  • Video title: {video_title}
  • Frame size: {width}x{height} @ {fps}fps (e.g. 1920x1080 @ 30fps)
  • Background: {background_description — footage, still, or solid color}

Labels

  • Center label: {center_label} (2–5 words — the one thing to remember)
  • Top-left corner: {corner_top_left} (1–3 words)
  • Top-right corner: {corner_top_right} (1–3 words)
  • Bottom-left corner: {corner_bottom_left} (1–3 words)
  • Bottom-right corner: {corner_bottom_right} (1–3 words)

Layout

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

Pro — les 511 compétences

Vous aimerez aussi