workflow video prompts

26 个提示词
GPT Image to Seedance Workflow Seedance 2.0
鼠标悬停播放预览

GPT Image to Seedance Workflow

15-second cinematic continuation shot, alternating slow-motion and real-time sequences in bookend pattern — slow-motion opening, chaotic real-time middle section, slow-motion finisher. Brutal close-quarters physical combat, no powers, no energy weapons, no time-freeze visuals, no stylized martial arts posing. Arri Alexa 65 with 40mm anamorphic lens shot at variable framerate (24fps real-time, 120fps slow-motion conformed to 24fps timeline), shallow depth of field, aggressive horizontal lens flares, slight anamorphic edge distortion, subtle gate weave during impacts. Bleach bypass color grading, desaturated mid-tones, crushed blacks, blown highlights, silver retention. Harsh sun-bleached canyon palette with pale limestone cliffs, oxidized copper mechanical internals, rust-coated debris, dried blood tones, deep crimson hydraulic fluid accents glowing subtly beneath fractured armor seams. CHARACTERS (heavily battle-damaged): — CRIMSON SENTINEL: towering white ceramic-plated combat android with cracked armor plating, deep gouges across chest and helmet, exposed copper pistons and torn hydraulic tubing visible through torso gaps, red visor flickering intermittently, left shoulder partially missing, right hand trembling from servo damage, crimson hydraulic fluid dripping onto sand, ceramic plating covered in claw marks and dust abrasion. Defensive combat stance constantly recalibrating through damaged joints. — UNKNOWN REPTILIAN CREATURE: tall sinewy biomechanical reptilian humanoid composed of dark oxidized flesh, exposed tendon-like tendrils, wet fibrous musculature wrapped around skeletal frame, elongated claws fractured and bloodied, jaw partially split exposing layered teeth, spine protrusions flexing during motion, oily blood and dust caked across torso, aggressive feral movement driven entirely by momentum and instinct. ENVIRONMENT: dry canyon basin under brutal midday sunlight, towering limestone walls reflecting hard heat, drifting dust clouds, scattered rusted wreckage and broken mechanical debris partially buried in sand, shimmering heat haze across the ground, dry wind carrying fine sand particles through frame. CAMERA: slow-motion sections use floating orbital camera drifts with subtle handheld inertia, real-time sections use violent handheld tracking and rapid reactive reframing. Tight medium-close framing during impacts, low camera height near waist level to exaggerate mass and speed. Heavy micro-shake during collisions, stabilized drift during slow-motion bookends. BEAT 1 — SLOW MOTION 5x OPENING (0:00–0:04): The reptilian creature launches across the canyon gap in a massive forward leap, sand exploding beneath its feet in suspended particles. Tendrils whip backward through the air while saliva and dark blood trail from its jaws in slow arcs. The CRIMSON SENTINEL pivots defensively, one damaged foot grinding into rock as hydraulic pistons compress under stress. The robot raises one fractured forearm shield-first while its other arm retracts near the torso preparing a counter. Sunlight streaks across cracked ceramic plating as floating dust drifts through shallow depth of field. The reptile’s claws extend inches from the robot visor as both figures converge mid-frame. BEAT 2 — REAL-TIME (0:04–0:06): Instant full-speed collision. The reptile slams into the robot’s guard with a deafening metallic crack. The robot absorbs impact, rotates its hips, drives a brutal knee strike into the creature’s ribcage, then immediately follows with a savage spinning backhand across the reptile’s skull. Teeth fragments, hydraulic fluid, and blood spray across frame. Camera violently whip-pans with impact momentum. BEAT 3 — REAL-TIME (0:06–0:08): The reptilian creature recovers instantly and lunges upward onto the robot’s torso, claws digging deep into cracked chest armor. Ceramic plates tear free as exposed internals spark mechanically. The added momentum drives the CRIMSON SENTINEL backward onto the canyon floor. Heavy body impact sends dust and pebbles outward in a shockwave across the sand. BEAT 4 — REAL-TIME (0:08–0:10): Ground exchange. The reptile repeatedly rakes claws across the robot’s faceplate, carving fresh gouges into the visor housing. The robot grabs the creature’s jaw with both hands, fingers crushing into wet tissue while servos scream under load. The reptile snaps violently inches from the robot’s visor while both bodies grind through dirt and shattered stone fragments. Broken armor shards scrape across the lens foreground. BEAT 5 — REAL-TIME (0:10–0:11.5): Both combatants stagger upright simultaneously. The robot’s damaged arm hangs partially limp while leaking crimson fluid down its fingers. The reptile’s jaw hangs dislocated and twitching. They stand chest-to-chest at close range breathing heavily — mechanical ventilation against wet animal snarling. Heat haze distorts the air between them while canyon wind carries drifting dust through the narrow space. BEAT 6 — SLOW MOTION 6x FINISHER (0:11.5–0:15): Both fighters attack simultaneously. The reptilian creature swings a full-force claw strike toward the robot’s head while the CRIMSON SENTINEL drives a crushing armored punch directly into the creature’s throat and jawline. Impact occurs mid-frame in ultra slow motion. Ceramic fragments, teeth, blood mist, sand particles, and crimson hydraulic fluid erupt outward in layered detail. The reptile’s flesh compresses around the robot’s fist while the robot visor fractures from the claw impact, red optics flickering violently. Camera slowly pushes inward during suspended destruction as debris rotates through sunlight. Final freeze frame on mutual impact with both faces inches apart, bodies twisting from collision force. Hard cut to black at exactly 15 seconds. DIEGETIC SOUND DESIGN ONLY, NO MUSIC, NO SCORE, NO STINGERS: — 0:00–0:04 (slow-mo): heavily lowered-pitch tendon stretching, deep metallic servo groans, elongated air displacement from leap, hydraulic creaking with extended tail reverb, sub-frequency impacts emphasized beneath canyon wind. — 0:04–0:06 (real-time): explosive metal-on-bone collision, sharp ceramic cracking, dense knee impact thud, violent backhand crunch, rapid sand scatter underfoot. — 0:06–0:08 (real-time): claws piercing armor plating, servo overload screech, heavy body slam into canyon floor, debris shower bouncing across stone. — 0:08–0:10 (real-time): scraping claws against ceramic shell, jaw snapping inches from microphone perspective, grinding stone friction, strained hydraulic pressure release. — 0:10–0:11.5 (real-time): mechanical breathing vents, wet guttural snarls, fluid dripping onto sand, distant cicadas and dry canyon wind. — 0:11.5–0:15 (slow-mo): ultra-deep impact bass, stretched ceramic fracture sounds, slowed bone crunch, elongated fluid spray textures, massive low-frequency resonance with extended reverb tail. — continuous ambient bed: dry canyon wind, distant cicadas, shifting sand, subtle rock debris movement, heat-driven environmental creaks. Photorealistic, IMAX-grade tactile detail. Slow-motion frames maintain hyper-detailed debris separation and visible material deformation. Real-time combat remains violently kinetic but readable on every impact frame. No text, no logo, no watermark.

workflow gpt-image cinematic storyboard
作者 Kang Ai
2026-05-08
Storyboard Animation Workflow with Seedance 2.0 Seedance 2.0
鼠标悬停播放预览

Storyboard Animation Workflow with Seedance 2.0

Part 1 @ Image1 as the storyboard reference. Strictly follow Storyboard 1: The Sleeping Beacon. K-17, a small maintenance robot with matte off-white ceramic shell, brushed bronze joints, single cyan camera eye, scratched glass chest light, tiny antenna, and worn red shoulder stripe, wakes inside an abandoned lunar lighthouse outpost. Maintain the exact same robot design in every shot. Stylized 3D animation, Blender render feel, warm highlights cold shadows, tactile scratched metal and dusty glass materials. No brand logos, no readable screen text, no camera gear visible. Scene 1: Wide shot of a forgotten lunar lighthouse half-buried in moon dust, Earth small in the sky. Slow dolly push over 3 seconds, eye-level horizon locked. Narrator: “On a silent moon outpost, one little machine still remembered its job.” Scene 2: Medium close-up of K-17 waking in a dusty maintenance bay, cyan eye flickering on. Locked-off camera, shallow depth of field. Narrator: “For years, K-17 guarded a beacon no one answered.” Scene 3: Over-shoulder shot of K-17 watching the tower lens fade above, no readable screen text. Slow tilt up from K-17 to the dark lens. Narrator: “Then one night, the signal died.” Scene 4: Close-up of K-17 uncovering a cracked amber power core beneath broken panels. Slow push from close-up to tighter close-up, reflections shifting across glass. Narrator: “But beneath the dust, it found one last spark.” Scene 5: Wide vertical-feeling shot of K-17 standing at the base of a huge spiral staircase, holding the heavy core. Low angle, slow pull back revealing the scale. Narrator: “So it began to climb.” Scene 6: Final wide shot, tiny K-17 climbing alone into the dark tower. Camera remains steady, beacon chamber above barely glowing. Tone: quiet, emotional, lonely, hopeful. Audio: low lunar wind, soft metal creaks, faint synth pulse, delicate piano notes. No on-screen text.

animation storyboard workflow tutorial cinematic
2026-05-07
GPT Image + Seedance 2 Character Card Dance Workflow gpt-image-2 Seedance 2.0
鼠标悬停播放预览

GPT Image + Seedance 2 Character Card Dance Workflow

用人物角色卡制作跳舞视频,只需两步,三分钟搞定! 第一步:GPT image2制作人物不同角色和姿势的参考图 第二步:把生成的图给到Seedance 2,输入简短跳舞提示词,坐等生成,搞定! (详细跳舞提示词见作者评论区)

workflow dance character seedance gpt-image tutorial
2026-05-06

Happy Horse Cinematic Trailer

I gave Happy Horse from Alibaba a spin to make this cinematic trailer. Grab my @magnific Spaces workflow in the comments below. A few things I noticed after a few days of testing: - Reference images work like they do in Kling 3.0 and Seedance 2.0. Drop in characters, locations, objects, and you can build much more dynamic shots than start/end frames allow. - Audio is clean. English dialogue comes out clear. If you don't want a soundtrack, a simple "no music" in the prompt gives you SFX only. - 1080p looks great but tends toward sharp edges. I usually generate at 720p and upscale with Magnific Video Upscaler. - Got surprisingly far with simple prompts. - Happy Horse is around 75% of the price of Seedance 2 at 1080p. Fun trivia: "Ma" means horse in Chinese, which also happens to be Jack Ma's last name. Make of that what you will. Props to the team at @alibaba_cloud team!

happyhorse alibaba cinematic trailer magnific workflow
2026-05-06
iPhone Vlog Workflow with Seedance Seedance 2.0
鼠标悬停播放预览

iPhone Vlog Workflow with Seedance

请根据参考图片中的人物、关系、地点和旅行氛围,生成一段自然真实的旅行 vlog 视频。 视频要像用旧 iPhone 随手拍下来的私人旅行记录,而不是广告片、旅拍大片或电影预告。镜头语言要非常生活化:手持拍摄、轻微晃动、偶尔失焦、轻微运动模糊、自然变焦、边走边拍、随手转向、偶尔被风吹乱头发、人物没有完全准备好就被拍到。 请保持参考图中人物的长相、发型、穿搭气质和关系感一致,不要换脸,不要变成陌生人。人物之间要有真实互动:聊天、笑、回头、牵手、靠近、互相拍照、分享食物、边走边闹、看风景时自然靠在一起。不要让人物一直正对镜头摆拍。 整条视频按照参考图片的故事顺序自然推进:从旅程开始,到白天探索,再到最有情绪的风景时刻,最后进入夜晚或房间里的收尾片段。画面要有 vlog 的随机感和呼吸感,可以有一些一闪而过的路人、街景、车窗、食物、手部、背影、侧脸和环境细节。 整体质感:低对比度、轻微褪色、旧手机拍摄感、自然光、真实生活噪点。不要出现文字、水印、字幕、logo、相册网格或拼贴边框。不要把参考图做成静态照片滑动展示,而是要把它延展成真实发生过的旅行视频片段。

vlog travel workflow realistic mobile-style
2026-05-04
Fantasy Battle Workflow Seedance 2.0
鼠标悬停播放预览

Fantasy Battle Workflow

"Live action. Use @image1 to create a 15s cinematic HBO fight scene video. No text.."

battle fantasy ninja storyboard workflow
作者 George Wu
2026-05-03
Hermes Agent + ComfyUI 工作流自动化
鼠标悬停播放预览

Hermes Agent + ComfyUI 工作流自动化

Hermes Agent 有了平替 Lovart 的潜力! Hermes Agent 集成了 ComfyUI 能力,让 Agent 可以自动创建 ComfyUI 工作流,很多人可能想不到这个的价值有多大,我24年做图像视频生成平台的时候,就觉得 AI 调用 ComfyUI 来自动化构建工作流是绝对的大趋势,后来年底 lovart 就是利用了 claude 3.6 的巨大能力提升实现了这个目标,现在 Hermes 集成了 ComfyUI,也就有了平台 lovart 的潜力。 另外,不要小看自动化构建工作流的潜力😎

workflow automation ai-tool creative
2026-04-30
LeonardoAI Cinematic Workflow Seedance 2.0
鼠标悬停播放预览

LeonardoAI Cinematic Workflow

Generate a 15s cinematic sequence 🎬🔥 This is where it comes to life. Here’s the exact prompt I used 👇 An elegant performer inspired by a legendary moonwalk dancer, dressed in a sleek black outfit with hat, transforming into a diamond trophy through motion His moonwalk generates energy trails that evolve into abstract light and crystallize into a diamond sculpture Luxury cinematic: slow dolly, orbit shots, macro details, seamless transitions, final push-in Scene dark premium studio, black reflective floor, golden spotlight, floating particles Ultra high-end commercial, diamond reflections, golden light particles, smooth VFX transitions - Movement phase wide shot, smooth moonwalk, subtle light trails appear - Energy phase, camera closer, golden trails intensify, lingering in space body begins dissolving into light - Abstraction phase, camera orbit, body becomes swirling luminous ribbons, abstract silhouette forms - Form phase, energy condenses into faceted diamond structure limbs sharpen into reflective geometry - Design phase, full diamond figure locked in iconic pose light glints across surfaces Final shot, trophy on pedestal, slow push-in, luxury finish CUT

advertisement luxury workflow image-to-video cinematic
2026-04-29
Yoga Flow Pose Transition Animation Seedance 2.0
鼠标悬停播放预览

Yoga Flow Pose Transition Animation

Scene: One continuous shot - Hand-held YouTube-style video showing a woman performing a complete yoga vinyasa flow. Character: Use IMG1 as the character we're following in the scene. Character Motion: Use IMG2 to follow the motion and instructions exactly to complete her flow. The scene starts at IMG1. Direction: The woman stands in a peaceful orange yoga studio. The hand-held camera subtly pushes in on her as she begins her flow. It captures the moment in tight composition as she moves from mountain pose to upward salute. The camera begins to pull back out as she transitions from upward salute to swan dive and stays focused on her body as she transitions from swan dive into a forward fold. The camera begins to move right to a side profile as she plants her palms and steps back into the high plank position. The camera drops lower and pushes in towards her face for a focused close-up as she moves from chaturanga lower to upward-facing dog. The camera again pulls out, showing her full body as she transitions to the downward-facing dog position. Finally, she holds for a deep inhale and exhale in that position. Overall Tone: Her movements are all natural and fluid, controlled and directed, strong with natural body movements and subtle readjustments as she gets her balance. SoundFx: peaceful ambient hum as we hear her inhale and exhale in a rhythmic fashion with each passing pose. No music.

dance pose animation tutorial workflow
作者 Rory Flynn
2026-04-29
AI Filmmaking Hacks Part 3 Seedance 2.0
鼠标悬停播放预览

AI Filmmaking Hacks Part 3

Create a fight scene between char 1 and char 2 that takes place in Location 1 add in proper fight kind of lighting, the fight should follows the exact sequence and movements from steps 1–12 shown in img 1. The music should just be fight SFX. There should be no dialogue, text, or narration

filmmaking workflow storyboard fight-scene seedance
2026-04-28
Cinematic Summoning VFX Seedance 2.0 Seedance 2.0
鼠标悬停播放预览

Cinematic Summoning VFX Seedance 2.0

Use @[image1] as the character reference sheet. Maintain strong consistency in face, hair, outfit, silhouette and key accessories. Create a cinematic 15s multi-cut VFX showcase video in realistic 3D. The scene should have a realistic 3D look with high-end CGI quality, physically believable materials, cinematic lighting, volumetric atmosphere and strong depth. The character performs a summoning in a grounded real environment. Start with a calm setup shot. Then show the first trigger of the effect around the character or on the ground. The summon effect builds in clear stages, with energy gathering, symbols or forms activating, force rising and a visible presence emerging. Show the summoned entity, object or presence appearing gradually and clearly from the effect. The character must move naturally with the effect and feel like they are channeling, directing and controlling the summon. Avoid stiff posing. Use body language, gesture and reaction that match the buildup and force of the effect. Structure: - setup shot - activation detail shot - build-up shot - peak summoning shot - emergence shot - final hero shot with the summon fully present Realistic 3D, cinematic, high-end CGI, visually grounded, readable VFX, smooth shot progression, believable lighting interaction, strong atmosphere. No text, no subtitles, no watermark.

VFX cinematic summoning workflow
作者 aimikoda
2026-04-28
Celebrity Sighting Scenarios with Seedance 2.0 Seedance 2.0
鼠标悬停播放预览

Celebrity Sighting Scenarios with Seedance 2.0

Unsteady handheld phone footage shot from [SHOOTER VANTAGE / SHOOTING POSITION] at [TIME OF DAY], [LENS/PHONE PHYSICAL POSITIONING DETAIL], producing [PRIMARY OPTICAL ARTIFACT, e.g. a faint smear of condensation fog, a smudge of fingerprint oil, a streak of rain] across the [FRAME LOCATION, e.g. lower quarter, right edge, top third] of the frame and intermittent [LIGHT FLARE TYPE, e.g. glare blooms, lens halation, prismatic streaks] from the [DOMINANT LIGHT SOURCE] reflecting back against the shooter's phone camera, the image is flat, auto-white-balance toggling between [COLOR CAST 1, e.g. a cool blue cast] and [COLOR CAST 2, e.g. an orange-amber push] as [LIGHT MIXING SCENARIO, e.g. ambient street light mixes with interior glow], color entirely ungraded and slightly washed out. At 0s the camera swings erratically [DIRECTION, e.g. left-to-right, low-to-high, side-to-side] hunting for the subject, frame wildly off-center, catching [BLURRED ENVIRONMENTAL DETAILS visible during the search] before the autofocus locks momentarily on [SUBJECT 1: physical build, age range, hair, distinguishing features] [SUBJECT 1 POSITION/LOCATION IN SCENE], [SUBJECT 1 WARDROBE: detailed top-to-bottom clothing description with fabric, color, fit, accessories, and posture/demeanor]. [OPTIONAL SUBJECT 2: positioning relative to subject 1, full physical description, wardrobe, body language toward subject 1, exuding [RELATIONSHIP/INTERACTION DYNAMIC]]. At 2s the autofocus drifts off [the pair / the subject] and [BACKGROUND ELEMENT] sharpens while [the subject(s)] go soft and blurry, the camera operator [SHOOTER REACTION, e.g. whispering urgently off-mic, cursing under breath, holding their breath] before it snaps back to focus at 4s with a visible hunting pulse. [INTERRUPTION 1, e.g. a pedestrian, a server, a passing vehicle] [crosses / passes / blocks] the frame at 5s, briefly obscuring [the subject(s)] behind [OBSCURING ELEMENT], the shooter dipping and angling to reacquire them through [SHOOTING MEDIUM, e.g. the glass, the foliage, between parked cars]. At 6s [INTERRUPTION 2, e.g. a camera flash from another phone further down the sidewalk, a passing headlight wash, a neon sign flicker] [REFLECTION BEHAVIOR] creating a hot chromatic flare across the [FRAME EDGE], cyan and red fringing visible on the high-contrast [HIGH-CONTRAST EDGE DETAIL]. [SUBJECT NAME/IDENTIFIER] [REACTION VERB, e.g. glances, flinches, turns] briefly toward [SHOOTING MEDIUM] at 8s, [REACTION DETAIL: expression shift, body movement, instinctive shielding gesture] before [returning to conversation / stiffening / leaning back into shadow]; the shooter holds completely still for two seconds, barely breathing. At 10s the phone [CAMERA SHIFT, e.g. dips, tilts, drifts] as the operator shifts weight, the frame [FRAMING ERROR, e.g. cutting off both subjects at the shoulders, clipping a head, losing them entirely] momentarily, [OPTICAL ARTIFACT EVOLUTION, e.g. the window fog smearing the bottom edge thicker], before rising again to reframe them, composition still imperfect, slightly tilted, [SUBJECT BODY PART] clipped by the frame edge. [AMBIENT AUDIO BED, e.g. distant traffic noise, low restaurant chatter, lobby murmur] bleeds through the audio throughout, [MICROPHONE INTERFERENCE, e.g. wind buffeting the microphone in a low rhythmic thump, fabric rustle as the operator shifts], faint [BACKGROUND VOICES OR SOUNDS] from [SOURCE, e.g. other pedestrians on the sidewalk, nearby tables], and at 12s [HALF-AUDIBLE EVENT, e.g. an excited whisper from someone just off camera says something unintelligible], followed by [SECONDARY AUDIO EVENT, e.g. the rapid-fire click-burst of a DSLR shutter from nearby, a car door slamming, a phone notification chime]. At 14s [ENVIRONMENTAL DETAIL, e.g. the interior light flickers slightly, possibly a server passing] and the autofocus hunts one final time before the clip ends at 15s with the frame still imperfectly held on [the subject(s)] through the [SHOOTING MEDIUM CONDITION, e.g. foggy reflection-streaked glass, leaf-broken sightline], [SUBJECT QUALITY 1] and [SUBJECT QUALITY 2] both visible but never cleanly captured.

image-to-video workflow celebrity scene-composition photorealistic
作者 maxescu
2026-04-28
Rita AI GPT-Image Seedance 2.0 Animation seedance
鼠标悬停播放预览

Rita AI GPT-Image Seedance 2.0 Animation

Using this character image as a reference, create a close-up live character animation. [Add your character action design here.]

animation character workflow
作者 underwood
2026-04-27
Seedance 2 Dance Choreography Workflow Seedance 2.0
鼠标悬停播放预览

Seedance 2 Dance Choreography Workflow

Create a seamless 15-second dance choreography video based on the uploaded reference image. Use the female dancer from the reference image as the main character. GPT Image 2.0 prompt for reference sheet: [STYLE] monochromatic grayscale illustration, 3D rendered character, clean instructional reference sheet, white background, comic-style cell grid layout, technical diagram aesthetic [LAYOUT] 4x4 grid layout, 16 panels total, each panel separated by thin black border lines, numbered cells from 1 to 16, consistent panel size [CHARACTER] young female dancer, athletic build, ponytail hairstyle, crop top and baggy pants, sneakers, same character in all panels [PANEL STRUCTURE - per cell] top-left: bold number badge + Korean title text center: full-body character pose illustration bottom-left: Korean description text (3-4 lines) overlay: directional arrows indicating movement direction [ARROWS / MOTION INDICATORS] curved arrows, straight arrows, circular rotation indicators, placed around the character to show movement flow and direction [RENDERING STYLE] high detail 3D sculpt style, soft studio lighting, subtle shadows, no color, grayscale shading, clean linework, game concept art quality [NEGATIVE] no background scenery, no color tones, no extra characters, no cluttered backgrounds

dance choreography workflow tutorial
作者 ai_gezgini
2026-04-26
TapNow Seedance Cinematic Workflow Seedance 2.0
鼠标悬停播放预览

TapNow Seedance Cinematic Workflow

been testing a different workflow lately using tapnow. what makes it interesting is how it structures the entire process from idea → visuals → final video. instead of jumping between tools, you can actually build everything in one flow and refine it step by step like a real production pipeline. for the visuals, i'm using seedance 2.0 which is currently one of the strongest models for photoreal, human-centered video. but quick note — seedance 2.0 is currently only available in selected regions and requires a verified corporate email to access. still, the direction is clear: AI video is moving from "generation" → into "directing". also, they just launched a global challenge called "10,000 Parallel Universes" with a $200K prize pool. if you're exploring cinematic AI workflows, this is actually a good place to test ideas and push concepts further.

cinematic workflow urban creative realistic
2026-04-13
Seedance 2.0 Storyboard to Video Seedance 2.0
鼠标悬停播放预览

Seedance 2.0 Storyboard to Video

Use the references with strict priority and role separation:[REF_STORYBOARD] = primary guide for shot order, framing, timing, composition, and scene progression. Follow it strictly.[REF_GIRL_MODEL] = character identity anchor for the teenage girl. [REF_PHOENIX_MODEL] = creature identity anchor for the phoenix. [REF_BACKGROUND] = environment and forest background anchor.Create a Japanese anime action scene of a teenage girl fighting two phoenix birds in a forest.Absolute priority:Follow the storyboard strictly for all shots and scene progression. Keep the girl design locked to [REF_GIRL_MODEL]. Keep the phoenix design locked to [REF_PHOENIX_MODEL]. Keep the forest and path consistent with [REF_BACKGROUND]. The girl: teenage anime girl, short black bob haircut, rust-orange headband, rust-orange wrap tunic, dark inner collar, cropped dark loose pants, agile fighter silhouette, determined expression, consistent face and costume in every shot.The phoenixes: two large mythic phoenix birds, copper-gold-orange plumage, pale chest, broad wings, long tail feathers, flame-like crest details, powerful talons, consistent design in all shots.The environment: lush forest path, tall trees, dense foliage, rocks, roots, grass, earthy ground, warm daylight, cinematic depth.Visual target: clean Japanese anime linework, polished final-frame rendering, crisp cel shading, dynamic camera angles, detailed forest backgrounds, strong action readability, finished anime film look.Important: storyboard is the main source for camera and shot progression no photorealism no 3D CGI no rough sketch look

workflow animation cinematic multi-shot storyboard
作者 VoxelPlot
2026-04-13
Seedance 2.0 + Topview Agent Seedance 2.0
鼠标悬停播放预览

Seedance 2.0 + Topview Agent

Original Action Short Film: It opens with a futuristic city with almost real movie texture, gradually transitioning to a high-energy two-dimensional action style. Characters chase, leap and confront at high speed between neon viaducts and high-rise buildings. The lens language is stable at first and then explosive; the materials transition from real metal and wetland reflections to exaggerated energy lines and dynamic painting sense, forming a visual effect of "integration of real movie sense and anime explosive sense". A strong hook in the first 2 seconds, with a stable main body, coherent actions, movie-level composition and light and shadow, real texture, epic sense, strong emotion, high-definition details, suitable for social media communication.

cinematic action workflow hybrid neon anime
作者 TopviewAI
2026-04-06
Multi Prompt Showcase Seedance 2.0
鼠标悬停播放预览

Multi Prompt Showcase

0–1.5s (Shot 1): Low-angle shot from the floor. A wooden chair drags slowly across the room on its own, stopping abruptly. Shadows stretch unnaturally along the walls as a door creaks open in the background. VFX: shadow stretch, dust particles SFX: wood scrape, door creak 1.5–3s (Shot 2): Crane shot rising. A woman stands frozen, breath shaking, as a dark silhouette forms behind her in the corner, barely visible in the dim light. VFX: shadow forming, light flicker SFX: low whisper, heartbeat 3–4.5s (Shot 3): Pull-out shot. The lights flicker violently, revealing the figure closer each time, its head tilted unnaturally, limbs slightly distorted. VFX: flicker strobe, glitch effect SFX: electrical buzz, sharp sting 4.5–6s (Shot 4): High-angle shot. Objects begin to lift slowly into the air, then slam violently against the walls as the room darkens completely. VFX: levitation, impact shock SFX: thuds, glass shatter 6–7.5s (Shot 5): Close-up push-in. The woman whispers a prayer, eyes filled with tears, as a hand suddenly grips her shoulder from behind. VFX: sudden shadow pop, micro shake SFX: whisper echo, grab hit 7.5–9s (Shot 6): Tracking shot. She is dragged backward across the floor into darkness, leaving scratch marks as her fingers claw at the wood. VFX: floor drag marks, motion blur SFX: dragging, scream echo 9–10.5s (Shot 7): Dolly shot circling. The room twists subtly, walls bending inward, the silhouette now fully visible in the center, watching silently. VFX: distortion warp, subtle rotation SFX: low drone, breathing 10.5–12s (Shot 8): Low-angle shot. The entity suddenly lunges forward unnaturally fast, filling the frame with a distorted face. VFX: speed warp, face distortion SFX: jump scare hit, scream burst 12–13.5s (Shot 9): Low-angle tracking shot. Silence drops. The room appears empty again, objects still, as if nothing happened. VFX: fade reset, dust settle SFX: silence, faint ringing 13.5–15s (Shot 10): Facial close-up. Behind the woman, the entity slowly reappears, inches from her face, eyes hollow and unblinking as it whispers. VFX: shadow reveal, eye darkness SFX: whisper close, breath

tutorial workflow tool multi-shot cinematic
2026-04-04
Action Storyboard Workflow Seedance 2.0
鼠标悬停播放预览

Action Storyboard Workflow

3x3のストーリーボードをリファレンスにしてアクションシーンが作れます オムニリファレンスがないですがこれはかなり使えそう さらに他では弾かれるレベルの実写の顔でも通っちゃいます

action tutorial workflow cinematic video
作者 とうや
2026-04-04
AI Video Workflow Seedance 2.0
鼠标悬停播放预览

AI Video Workflow

画面の 左側が「入力・設定エリア」 右側が「制作エリア」 【左側】入力・設定エリア ここに、人間が「総監督」として、作品の方向性を入力していきます。 ユーザーの抽象的なリクエスト(例:戦うシーン)に対し、AIディレクターが即座に制作プランを提示。 「キャラクター」を入れると具体的なワークフローが自動で組まれます。 動画の長さ、アスペクト比などをワンクリックで選択。 セリフ・スタイルの入力は、プロンプトではなく日常的な言葉での指示だけでOK! 【右側】制作エリア 左側での決定に基づき、複数のAIエージェントが並列・連携して自動で動き生成してくれます。 脚本 (最上段): 入力された「戦うシーン」から、シナリオと演出を生成。 キャラクター設定 (二段目): 脚本に基づき、三面図を含む詳細なキャラクターシートを作成。 環境設定 (三段目): 戦いの舞台となる背景を構築。 最終出力 (最下段): すべての設定を統合。 複数のカットがストーリーの流れに沿って一気に書き出されます。

tutorial workflow tool japan meta
作者 AICO_fly
2026-04-03
Sneaker Product Ad Seedance 2.0
鼠标悬停播放预览

Sneaker Product Ad

Extreme close-up on the toe area. thousands of glowing white fibers burst out, twist and interlock, gradually forming the front shape of a sneaker. The surface develops a detailed 3D woven texture. Soft glow, high-tech feeling, sharp focus, shallow depth of field.

product workflow commercial macro fashion
作者 el.cine
2026-04-01
Character Consistency Workflow Seedance 2.0
鼠标悬停播放预览

Character Consistency Workflow

I finally cracked character consistency in Seedance 2.0 • Works with real humans • Works with fictional characters • Train once → generate infinitely No hacks, just a solid workflow 🔥 Now open-sourced in Open-Higgsfield-AI

tutorial technique workflow
2026-03-29
Cinematic Workflow Template Seedance 2.0
鼠标悬停播放预览

Cinematic Workflow Template

[CINEMATIC SETUP] [Film stock / lens / aperture / camera behavior]. [Color grade]. [Lighting source]. [Atmosphere]. [Audio: sound FX only / no music]. Face stable, no deformation. [@ image1] is [role]. [@ image2] is [reference for X]. 0-1s: [Shot type]. [@ image1] [precise action]. [Camera move]. [Sound.] 1-3s: [Shot type]. [Action sequence with physics]. [Camera behavior]. [Sound FX.] 3-5s: ... 13-15s: [Closing shot]. [Final action / movement]. [Sound fade or impact.]

workflow tutorial cinematic template
作者 0xInk_
2026-03-26