Craig Watson

Loved this first episode of "The Gray Matter" from @aizelskystudio videee.com/video/258 Such a talented AI video creator.

Time to go a bit insane.

Seedance 2.0 on Higgsfield AI Full open-sourced prompts & assets below: @Image1 is the performer — preserve her exact identity: cornrow braids into long dark curls, silver chain-drop earring, gold pendant necklace, washed-green denim corset top, glossy lips. @Audio1 is the finished master track — the only audio; no invented music, no new vocals. SHE RAPS THE VOCAL ON CAMERA — PRECISE LIPSYNC IS THE TOP PRIORITY. Mouth articulates every syllable of @Audio1 exactly on time; face visible and sharp through every vocal line, no cutaways mid-word. LIPSYNC MAP: 0.0–0.5 instrumental. 0.5–2.8 "I'm standing on the edge / Say it with your chest / Or keep it on the deck" + "Hey!" 3.2–6.7 "I walk in, whole room gets tense / I don't need luck, I'm the consequence / If you really want to test my intent / Come correct, come correct or get bent" + "Woo!" 7.5–13.7 same hook verbatim second time, escalated. 14.5–15.0 instrumental hold. Music-video route: dance/performance, NYC caged basketball court. Director thesis: queen of the cage — she raps at the center of a chain-link basketball court while her girls run formations around her. Black girlie hustle vibe. Visual world: outdoor NYC basketball court wrapped in tall chain-link fence, faded court paint, brick housing blocks and fire escapes behind, late golden-hour sun throwing fence-diamond shadows across the asphalt. 4 background dancers: Black girls styled sharp — matching cropped white tanks, baggy cargo denim in different washes, bamboo hoops, colorful silk scarves tied on heads, clean Timberlands and Air Force 1s. Palette: asphalt grey, golden sun, brick red, denim, gold jewelry. No neon, no particles. Shot flow (crash zooms through the chain-link): 0–0.5s static wide THROUGH the chain-link fence: she stands at center court, four girls scattered frozen. 0.5s CRASH ZOOM IN through the fence diamonds to her face on the first word — she raps dead center, fence shadow patterning the asphalt around her. 2.8–3.2s on "Hey!" all four girls snap into a diamond formation around her on one beat. 3.2–6.7s she raps strutting the free-throw line, camera dollying laterally inside the cage; CRASH ZOOM punches on "tense" and "consequence"; the girls hit unison hip-hop choreography — chest pops, scarf-flick accents, synchronized bounce — locking freezes on every snare; finger to lens on "come correct". 7.5s CRASH ZOOM IN on the kick to a low-angle close-up against the sun — second hook doubled, braids backlit, jaw snapping consonants. 10–13.7s fast orbital around the whole formation as the girls run a rotating carousel around her rapping at center; two more crash zooms slamming in on snare hits; fence shadows strobe the ground as the camera circles. 13.7–15s on "Woo!" CRASH ZOOM OUT through the fence to the full wide: she's frozen arm high at center, girls dropped into four squat freezes around her, hold. Loopable. Performance rules: feminine swagger, athletic bounce, immaculate diction; only she raps, girls never mouth words. Continuity: same five women, same court and light. Audio intent: @Audio1 only, mouth locked to it; faint ball-bounce echoes and city hum under the track. Quality bar: expensive NYC rap video, real film texture, no AI gloss. Higgsfield Originalsの制作プロジェクトがオープンソース化され、完成映像だけでなく、実際に使用されたプロンプトや参照画像・音声、モデル、画質、解像度などの生成設定まで確認できるようになりました。 今回選んだのは、NYCのバスケットボールコートを舞台にしたラップパフォーマンスの1ショット。 リップシンクを最優先にしながら、クラッシュズーム、カメラの周回、ダンサーのフォーメーション、光やフェンスの影まで、15秒の中にかなり細かく設計されています。 完成映像と実際のプロンプトを見比べることで、「どの指示が、どの映像表現につながっているのか」を確認できるのが面白いところ。 プロンプトのコピーやクリップのダウンロードもできるので、再現を試したり、自分の制作に合わせて構成や演出を研究したりと、AI映像を作る側としてかなり参考になる内容です。

i love how good seedance 2 is in animation hope 2.5 comes faster #AIart️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️️

本日から1週間、 渋谷の街頭ビジョンで、 私たなか の 作品が放映されます📢📢 まさか自分の作品が、 東京の中心で流れることになるとはｗ 改めて、川津さん、 貴重な機会をいただきありがとうございます🙏

@MarinSynArt @whorange__ Yeah this is dope!

@whorange__ Such a great animation! The action is so high paced and works so well with this style!

Two assassins. Two completely different philosophies. One relies on precision, speed, and a storm of razor-sharp knives. The other answers with relentless suppressive fire, constantly pushing forward through the chaos. --sref 742788275 Really love how this monochrome comic-book aesthetic amplifies the intensity. High contrast ink, aggressive linework, and nonstop Hollywood style action. Made with Seedance 2.0 on Showrunner Studio @fableshowrunner

@carolletta Here’s mine x.com/jaynwabueze/st…

They said attention is the most valuable currency. She collected enough to build a throne. Seedance on @kinovi_ai

Kade, the wolf🐺. Made with Midjourney -> Seedance 2.0 + Suno. 10 Seedance 2.0 generations of 15 seconds each, 9 of which were used. No character sheet, but a reference image from the first generation.

New Music Video out for MNTRA’s remix of D.A.N.C.E. From Justice. Proud to be a creative partner of @Hailuo_AI.

We think we're the most advanced civilization that's ever existed, so did the last one....and the one before that. My entry for the @OiiOii_AI Into the Paradox contest. #OiioiiAiParadox

Can you prompt better? I can't.

🐮☁️👁️‍🗨️

Morning Coffee & Study Setup Vlog Seedance 2.0 on Hailuo_AI Prompt : CAMERA / LOOK: Handheld mini DV camcorder footage filmed by the subject herself. Slight hand shake, occasional focus hunting, imperfect framing, natural zoom adjustments, soft tape-like image quality, subtle grain, realistic auto-exposure shifts from morning window light. Natural skin tones, mild motion blur, authentic consumer camcorder aesthetic. STYLE: Cozy study-desk vlog with gentle ASMR elements. Relaxed pacing, minimal dialogue, candid moments. Focus on satisfying sounds: coffee grinder, pouring water, page flipping, pencil case zipper, sticky notes peeling, mug clinks, keyboard taps. SUBJECT: Young woman in her early 20s, messy bun, oversized knit sweater, glasses, minimal jewelry. Calm, focused energy before a study session. SETTING: Small home study corner on a weekday morning. Soft window light, wooden desk, plants, open notebook, laptop, mug of coffee, quiet atmosphere. STORYBOARD: → (2s, propped medium shot) Places camera on desk, sits down. "Starting my morning study routine." → (2s, overhead shot) Grinds coffee beans, pours hot water over filter. Only brewing sounds. → (2s, close-up) Opens notebook, flips through pages, uncaps a pen. → (2s, handheld desk shot) Arranges sticky notes and highlighters. "I need to actually use a color system this time." → (2s, detail shot) Places headphones on, adjusts laptop screen. No dialogue. → (2s, medium shot) Sips coffee while reviewing a to-do list. "Okay, let's see what's on the list today." → (2s, macro shot) Pencil case zipper, pen selection, eraser placed on desk. → (2s, propped shot) Opens laptop, types a few lines, soft keyboard clicks. → (2s, warm shot) Looks at the cozy desk setup, satisfied smile. "This is my favorite time of day." → (1.5s, final shot) Reaches toward camera. "See you after class." Hand covers lens. AUDIO NOTES: Natural room ambience grinder, pouring water, page turns, pen clicks, keyboard taps should be clearly audible. Dialogue quiet and casual. REALISM NOTES: Authentic body language, natural blinking, genuine small smiles, occasional pauses, imperfect framing, focus breathing, morning light shifts. Should resemble a genuine personal study vlog on a consumer camcorder, not a commercial production.

This Seedance 2 prompt is an absolute banger for post-apocalyptic sci-fi. ⚡️🔥 Honestly, the atmosphere of this concept goes really hard. Made in Hailuo PROMPT below 👇 A lone ex-soldier in damaged body armor with a cracked helmet and glowing wrist device + 0–4s: moves silently through a dead industrial zone while drones scan the ruins; - enemy machines detect him, explosions erupt, he fights through robots with EMP blasts, shotgun shots and brutal hand-to-hand takedowns; - activates a final pulse that drops every drone from the sky, standing alone beneath falling sparks as the camera pulls back over the destroyed factory Post-apocalyptic sci-fi action, sparks, smoke, practical explosions, dynamic tracking, final silence after chaos. Want me to share more sci-fi prompts like this? 👇

A small scene from my first horror episode. 👁️ I've been experimenting with a darker, more unsettling style lately, and I'm really enjoying where it's going. --sref 802244360 Full episode coming soon 🖤 Made with Seedance 2.0 on Video Studio @capcutapp #capcut #seedance2

Somewhere past halfway, it became real.

Happy to announce that “120 seconds“ is 90% done ✅ Creating this short film has thought me a whole lot about AI Filmmaking, and I can’t wait to share it with you guys.

Samurai fight scene. Tool: @dreamina_ai Seedance 2.0 prompt Cinematic multi-shot sequence, one continuous duel, no unmotivated cuts, photorealistic, 8K ultra-high-definition, hyperdetailed, 35mm film quality, professional color grading, sharp focus, fine film grain, depth of field mastery, tracking-to-handheld fluidity, 180° shutter motion blur. WARRIOR ONE: a lean older swordsman, weathered face, grey-streaked topknot, faded indigo kimono with a torn sleeve, katana held low in a relaxed grip, calm and economical in movement. Same appearance throughout, no drift. WARRIOR TWO: a younger, broader swordsman, shaved sides with a dark top-knot, black armored haori over a crimson under-layer, a slightly longer blade held high in an aggressive stance, ferocious and explosive in movement. Same appearance throughout, no drift. LOCATION: a narrow snowy mountain pass at dusk, sheer grey rock walls on either side, deep snowdrift underfoot, thin dead pine trees, wind driving loose snow sideways, a bruised violet-grey sky, faint warm light bleeding from the last sliver of sun on the horizon. COLOR 60:30:10 — 60% cold slate-grey and white snow / 30% deep violet-blue dusk shadow / 10% accent of the dying warm orange horizon light and a flash of crimson from Warrior Two's under-layer. PHYSICS: real blade weight and momentum, boots break through snow crust with each step, snow kicks up and drifts in the wind on every strike, cloth and hair whipped by wind and motion, sparks on blade contact, breath steaming heavy in the cold. 0–3s — THE CLOSE: wide low shot, both warriors circling at a distance in the pass, snow driving sideways between them. Warrior One's calm stance against Warrior Two's coiled aggression. They surge together, blades clearing in the same instant. 3–5s — FIRST CLASH: tight tracking shot as the swords meet in a shower of sparks, both warriors pivoting past each other, snow exploding from their footwork, a beat of stillness as they land facing opposite directions. 5–8s — THE EXCHANGE: fast tracking shot circling the pair as blades trade in a rapid three-strike combo, blocked, parried, blocked again, snow and steam thrown with every impact, Warrior Two forcing Warrior One back toward the rock wall. 8–9s — THE OPENING: Warrior One slips a strike by a hair, time begins to ramp as his blade angles in toward Warrior Two's guard. 9–12s — THE MOMENT, slow motion 120fps: near standstill as the blades lock edge to edge, sparks frozen mid-scatter, snow suspended in the air around them, both warriors' faces taut with strain, breath visible and hanging, the dying orange light catching the steel. 12–15s — THE BREAK, snap back to 24fps: time cracks back, Warrior One twists free and disengages with a low sweeping strike, Warrior Two staggers back a step into the snow, both warriors resetting into stance, chests heaving, blades still raised, the wind howling between them as the shot holds wide. AUDIO (no music): howling mountain wind, boots crunching through snow, ragged cold breath. On each clash, sharp ringing steel-on-steel. During slow-mo, sound drops into a muffled vacuum, only a deep resonant blade-hum and both warriors' straining breath. At 12s a sharp metallic ring and gust of wind as full sound crashes back. FORBIDDEN: no CGI energy blades or glowing effects, no floating or weightless swords, no morphing or extra limbs, no design drift between warriors, no plastic CGI gloss, no cartoon proportions, no readable text or logos, no franchise or recognizable characters, no desaturated grading, no warping geometry. #DreaminaCPP