即梦AI肢体与动态人物穿搭
#肢体与动态#人物穿搭
中文翻译
主观视角:你正和最好的朋友闲逛,直到发现有人正在拍摄。
使用 Seedance 2.5 制作。 @Flovaai
提示词:创作一段 30 秒、1080 线逐行扫描、超写实的二十一世纪初数码家用录像,记录两名年轻韩国女性在首尔一处老街区的安静午后一起闲逛。
主要人物
人物甲:@ 图片 1,穿合身的粉蓝色短款上衣、宽松的奶油色睡裤式长裤、黑色运动鞋,戴简约银项链。
人物乙:二十岁出头,直发齐肩,用发夹夹住头发,宽大牛仔夹克内搭纯白短袖上衣,穿黑色紧身裤和穿旧的帆布运动鞋。
自然的皮肤纹理,淡妆,真实的身体比例。全程保持两人的身份、服装和发型完全一致。
环境
首尔一处老街区的安静住宅街角,午后过半。矮墙,附近有小吃推车或小摊,停放的自行车、电线杆,以及可坐下的路缘。环境普通且富有生活痕迹,无地标或品牌名称。
动作顺序
两位朋友一起走向一个街头小吃摊,笑着谈论之前就已开始的话题。她们买了装在纸杯里的辣炒年糕和鱼饼串。
她们走了几步,一起坐在低矮的路缘上,把纸杯稳放在膝盖上。
她们边吃边聊,对话与笑声大多听不清,或只能听到一部分——偶尔可能清楚听到一两句零散的话(“……不会吧,真的?”),但大部分都是自然的低声交谈,而非清晰的对白。
人物乙拿出一台便携式光盘播放器,递给人物甲一只耳塞。两人都戴上耳塞,轻轻随着节奏点头,仍在节拍间隙继续吃东西。
她们因某件事笑着靠向彼此,其中一人差点洒出杯里的食物。
人物甲发现镜头正对着她们,笑着说道:
“等等,你在拍我们吗?”
人物乙转头看过来,笑起来,迅速抬手伸向镜头。
她的手和手指突然遮住整个画面,录像切为黑屏。
摄影机
未经加工的二十一世纪初手持数码摄像机拍摄,画面晃动,自动对焦来回寻找焦点,颜色褪淡,曝光变化,运动模糊,数码细节柔软,偶尔发生无意的变焦。不采用精致的电影摄影。
声音
自然的街头环境声——小吃推车的煎炒滋滋声、远处的交谈声、自行车声、鸟鸣、共用耳塞中隐约漏出的闷闷的光盘播放器声音、脚步声和风声。主要是交叠的、只能部分听清的对话和笑声;除了前述两句,不要清楚的剧本对白。不要配乐。
连续性
两人手中的食物纸杯和串签保持一致。光盘播放器和耳塞始终由她们共用。不得出现物体消失、重复或瞬移。全程动作流畅、真实。
负面要求
计算机生成图像的质感、塑料皮肤、身份漂移、更换服装、扭曲的双手、多余手指、字幕、水印。
原帖
POV: you’re just hanging out with your bestie until you realize someone’s filming.
Made with seedance 2.5 @Flovaai
Prompt: Create a 30-second, 1080p ultra-realistic early-2000s DV home video of two young Korean women hanging out on a quiet afternoon in an older Seoul neighborhood.
MAIN SUBJECTS
Subject A: @ image 1 , fitted pastel-blue short top, loose cream pajama-style pants, black sneakers, simple silver necklace.
Subject B: early 20s, straight shoulder-length hair with a clip, oversized denim jacket over a plain white tee, black leggings, worn canvas sneakers.
Natural skin texture, minimal makeup, realistic proportions. Keep both identities, outfits and hairstyles perfectly consistent throughout.
SETTING
A quiet residential street corner in an older Seoul neighborhood, midafternoon. Low walls, a snack cart or small stall nearby, parked bicycles, utility poles, a curb to sit on. Ordinary and lived-in, no landmarks or brand names.
SEQUENCE
The two friends walk up to a small street food stall together, laughing about something already in progress. They buy tteokbokki and fish cake skewers in paper cups.
They walk a few steps and sit down together on a low curb, cups balanced on their knees.
They eat and talk, mostly inaudible or half-heard chatter and laughter — a stray phrase or two might land ("...no way, seriously?") but most of it is naturalistic murmur, not clear dialogue.
Subject B pulls out a portable CD player and offers Subject A one earbud. They both plug in, bobbing their heads slightly, still eating between beats.
They lean into each other laughing at something, one nearly spilling her cup.
Subject A notices the camera pointed at them, laughs, and says:
"Wait, are you filming us?"
Subject B glances over, laughs, and quickly reaches a hand up toward the lens.
The frame is abruptly covered by her hand/fingers, footage cuts to black.
CAMERA
Raw handheld early-2000s DV camcorder, shaky footage, autofocus hunting, faded colors, exposure shifts, motion blur, soft digital detail, occasional accidental zooms. No polished cinematography.
AUDIO
Natural street ambience — sizzling food cart, distant chatter, bicycles, birds, muffled CD player audio bleeding faintly from the shared earbuds, footsteps, wind. Mostly overlapping, half-audible conversation and laughter; no clear scripted dialogue except the two lines noted. No score.
CONTINUITY
Food cups and skewers remain consistent in each subject's hands. CD player and earbuds stay shared between them. No disappearing, duplicating or teleporting objects. Smooth, realistic movement throughout.
NEGATIVE
CGI look, plastic skin, identity drift, outfit changes, distorted hands, extra fingers, subtitles, watermark.
