▶ Style Capture · Storybook City Reel

Vancouver, World Cup Week — in a hand-drawn storybook

The FIFA World Cup opens June 11, 2026, and Vancouver is a host city. This reel follows Mei — black bob, mustard-yellow coat, sage-green crossbody — through opening week downtown: Gastown coffee, the Granville fan march, BC Place at golden hour, the fan-zone goal roar, and lantern-lit Chinatown. Every frame in one captured illustration style: flat 2D, ink outlines, blossom petals, everyday life is beautiful.

format
9:16 · 1080×1920
scenes
5 × 10s, all animated
style lock
verbatim style block
image → motion
gpt-image → ltx-q-i2v
music
instrumental spring jazz
The style block is the asset. One paragraph — transcribed from a reference reel — pasted verbatim into all five keyframe prompts is what makes five different Vancouver locations read as pages of the same book. Mei stays recognizable the same way: a name + a fixed wardrobe, restated in every prompt. No logos were drawn anywhere — generic flags, plain jerseys, an abstract pitch on the big screen.

Finished reel — the stitched short

All 5 scenes · 50.2s · 1080×1920 · every beat animated with ltx-q-i2v (10s each, motion-gated: visible character action between t=2s and t=7s, style must stay flat-2D) · a looped instrumental spring-jazz bed under the full reel, no silent tail.

Keyframes — one storybook page per beat, rendered on gpt-image

Each beat is a native 9:16 still from gpt-image (image_size:"portrait_16_9") — the full style block verbatim + the scene beat + Mei's name and wardrobe restated. Vancouver's landmarks are drawn into the style: the steam clock, the Granville marquees, BC Place's white crown, the Millennium Gate.

Mei with morning coffee by the Gastown steam clock
01 · MORNING
Gastown coffee
The steam clock puffs; 'MAPLE HOUR' café; World Cup bunting and small flags between the lamps. Mei warms her hands on the cup.
Granville Street fan march with flags and confetti
02 · MIDDAY
Granville fan march
Crowds in many nations' plain jerseys wave generic flags down Granville. Mei waves a small Canadian flag; confetti and petals drift.
BC Place stadium at golden hour, drawn storybook style
03 · GOLDEN HOUR
BC Place
The white-crowned stadium drawn naive-style, streams of tiny fans flowing in. Mei raises her phone for a photo.
Fan-zone plaza at dusk with a goal-moment confetti burst
04 · DUSK
Fan-zone goal moment
The big screen glows with an abstract match — tiny green pitch, little dots, no logos. Confetti bursts; Mei mid-cheer, arms up.
Chinatown Millennium Gate with red lanterns, Mei and friends with bubble tea
05 · EVENING
Chinatown lanterns
The Millennium Gate, red lanterns glowing, warm shop windows. Mei and two friends toast with bubble tea; petals settle.

The captured style block — pasted verbatim into every keyframe prompt

"Flat 2D hand-drawn storybook illustration, folk-art naive style. Clean uniform dark-brown ink outlines; flat color fills with subtle paper-grain texture; NO gradients, NO 3D shading, NO photorealism; at most minimal soft shadows. Warm muted retro palette: brick red / terracotta, cream / ivory, slate teal-blue, sage and olive green, mustard ochre, dusty rose pink, charcoal; warm off-white sky. Dense European-style streetscape: tall narrow brick townhouses with white-trimmed sash windows, dormers and chimneys; ground-floor shopfronts with dark hand-lettered signboards, window flower boxes, awnings; vintage cast-iron street lamps. Many small charming people at naive proportions (heads ~1/5 of body): vintage 1940s–60s attire — long wool coats, cloche hats, fedoras, scarves, satchels; rosy single-dot cheeks, simple dot eyes; everyone engaged in everyday actions (strolling, chatting at the café counter, cycling, carrying coffee or flowers). Busy but orderly composition with overlapping vignettes; larger figures in the foreground, street receding behind. Blossom trees frame the edges with confetti-dot foliage; petals drift through the air and dot the ground. Mood: cozy, nostalgic, gentle — everyday life is beautiful."

Tested verdict — honest, not inflated

5/5 style + cast lock · gpt-image + ltx-q-i2v verbatim style block
All five keyframes passed the style gate (flat-2D storybook per the verbatim style block, Mei in the same mustard coat + sage bag in every scene), and all five ltx-q-i2v clips passed the motion gate — frames at t=2s vs t=7s show visibly different action (sips, flag waves, the goal-moment confetti burst), not Ken-Burns drift. Final reel: 50.2s, 1080×1920, instrumental bed covering the full video (ffprobe: video 50.17s / audio 50.16s). Honest caveats: AI-rendered signboard text is decorative, not reliable; i2v motion in this flat style is gentle by design.

Pipeline trace — the cap chain this reel ran

Step 1 — script 5 beats · protagonist NAMED with fixed wardrobe ("Mei … black bob, mustard-yellow wool coat, sage-green crossbody") restated in EVERY prompt
Step 2 — keyframes 5 × gpt-image image_size:"portrait_16_9" — style block VERBATIM + beat → 5/5 · 608×1088 portrait gotcha: fal's gpt-image takes the size enum — pixel strings like "1024x1536" 422
Step 3 — motion (motion-gated) 5 × ltx-q-i2v resolution:"auto" · num_frames:241 · frames_per_second:24 · generate_audio:false → 5/5 clips · ~10s each · all passed the t=2s vs t=7s action gate (coffee sip · flag wave · phone raise · confetti-burst cheer · bubble-tea toast) gotcha: ltx-q-i2v IGNORES duration — pass num_frames; and "1080p" 422s (seedance is the opposite: requires "1080p") gate: frames at t=2s vs t=7s must show different poses/action; style must stay flat-2D
Step 4 — music music lyrics_prompt:"[Instrumental]" · "warm spring jazz, brushed drums, upright bass, light piano, cheerful café mood"
Step 5 — assemble + gate concat → 1080×1920 30fps yuv420p → music looped under the full reel (-stream_loop -1 + -shortest) → ffprobe: video 50.17s · audio 50.16s · 1080×1920 · Δ 0.01s

Honest notes — what to watch for

Signboard text is decorative, not dependable — keep it to one or two words ('MAPLE HOUR') and put anything load-bearing in post. Style drift is the #1 i2v risk: ltx-q occasionally re-lights a scene toward 3D or morphs the dot-eyes face — that's what the motion gate is for; reject and re-roll with "flat 2D illustration stays flat" emphasized rather than shipping a morph. Wardrobe is the consistency ceiling: name + restated wardrobe holds Mei at roughly 85–95% across five scenes; tiny accessories (the sage crossbody) drift the most, and a per-character LoRA is the 95%+ path if this becomes a series. Crowd scenes dilute the protagonist — always anchor the hero "in the foreground", or the i2v model animates the crowd and forgets the character action.

Reproduce this — the playbook is open

This reel was built from the storybook-city-reel playbook — bring any story (a trip, a week, a festival), keep the style block verbatim, name your protagonist, and the same chain renders it as a cozy illustrated reel. The style block is reusable across stories; that's the point.