AI 沙发换装趋势四套造型,一个旋转镜头

你躺在一张橙褐色沙发上。镜头升到你正上方,缓慢转动,然后快速旋转,每次旋转都会展示一套新造型。选择无人机旋转、街头风、正装风或宠物版。上传一张照片,生成 12 秒竖屏视频。

沙发换装是一种从上方拍摄的时尚视频:你躺在沙发上,镜头在你上方旋转,每次快速旋转都会隐藏剪辑并切换到新造型。AI 沙发换装趋势用一张照片生成所有四套造型,所以你无需换衣服。你只需要一张面部清晰的照片。

Also called 沙发时尚转场 · 旋转换装 · 旋转沙发换装 · 俯拍沙发写真

Example

沙发换装视频效果

第一个示例是我们自己的无人机旋转版本,使用 AI 生成的人物。第二个是这个趋势模仿的真实视频,使用真人模特拍摄,不是 Smew AI 的结果。

输入
我们的无人机旋转版本,使用 AI 生成的女性:镜头升到橙褐色沙发上方并旋转,每次旋转都会展示四套街头风造型中的下一套,配合视频的节拍。
Use this reference
@meshki 原始视频,10 秒
The original reel by @meshki (23 February 2024): a real model on a white curved sofa, filmed from straight above, in one outfit after another. This is the 4 outfits one sofa format that the AI trend copies.
Use this reference
Step by step

如何在 Smew AI 上制作沙发换装视频

全部在浏览器中运行。无需编辑应用,无需编写提示词,无水印。

  1. 添加你的照片

    点击"创建你自己的",添加面部清晰的照片。可以用相机拍摄或从相册中选择。

  2. 点击生成

    按钮会显示价格。Seedance 2.5 将你的照片放入沙发换装趋势视频中,逐镜头复制 12 秒视频,配有原视频的音乐。你无需编写提示词。

  3. 下载或分享

    视频将在几分钟内准备好,格式为 9:16 MP4。下载它,从手机分享,或稍后在"我的视频"中查找。

哪种照片效果最好?

你需要一张照片:街头风或正装风用你自己的照片,宠物版用你宠物的照片。

  • 人物:面部清晰、锐利且未遮挡,光线良好。半身照或腰部以上效果好;如果面部足够大能清楚看到,全身照也可以。视频大部分时间从上方俯视你,所以清晰的面部比你的衣服更重要。你的衣服会被四套造型替换。
  • 宠物:整只动物在视野中,面部和花纹清晰。每张照片一只动物。
  • 使用你自己的照片、你自己的宠物或已同意的成年人的照片。绝不使用名人、陌生人或儿童的照片。

沙发换装趋势起源于哪里?

The format is older than the AI version. On 23 February 2024 the fashion label @meshki posted a clip of a model lying on a white curved sofa, filmed from straight above, changing from one outfit to the next ("How MESHKI girls be lounging around"); it had 47,000 views on 4 November 2026. On 12 July 2025 @ellierosebakerr posted the same idea as "4 outfits one sofa".

The AI version adds the move that makes it hypnotic: the camera starts low at the foot of the sofa, rises to straight above, turns slowly, then whips into a fast spin, and the outfit changes inside the motion blur. A version with a man in four streetwear looks on a burnt-orange sofa was going round AI video sites as a community template by 28 October 2026. On 30 October @kevinbigj posted a Korean CapCut template called "sofa fashion transition" with the same camera move, and on 2 November an AI photo app started advertising a top-down couch photoshoot with four outfit changes.

It is early. On 4 November #sofaoutfit had 32 TikTok videos and 217,847 views, and #couchphotoshoot 18 videos.

  1. @meshki posts a model in a row of outfits on a white curved sofa, filmed from straight above; 47,000 views by 4 November 2026. TikTok

  2. @ellierosebakerr 发布"4 outfits one sofa",从上方展示一张沙发上的四套造型。TikTok

  3. 一个 AI 版本出现,使用旋转的俯拍镜头,一名男子穿着四套街头风造型躺在橙褐色沙发上,在 AI 视频网站上作为社区模板分享。

  4. @kevinbigj 发布名为"sofa fashion transition"的韩语 CapCut 模板,使用相同的升起和旋转镜头运动。TikTok

  5. 一个 AI 照片应用开始宣传带有四次换装的俯拍沙发写真。

Copy and paste

沙发换装提示词

你不需要这些:一键流程会为你运行它们,与趋势视频一起。[Image 1] 和 [Image 2] 是你的两张照片,[Video 1] 是视频复制的趋势视频。

Streetwear

1 photo: you, face clearSeedance 2.512 s · 480P · 9:16152 credits
Video prompt

Create a 12-second cinematic fashion editorial video featuring the same person wearing four different outfits, based strictly on [Image 1]. The person = the person in [Image 1], in EVERY shot: copy the exact face from [Image 1]: face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles exactly as in the photo. Take only the face, the hair, the skin tone and the body shape from [Image 1], never the clothes. The same real person in all four outfits, so their friends recognize them at once. Never morph. Reproduce each outfit's clothing, colors and layering exactly; never mix clothing between outfits. SCENE: a stylish vintage living room with a burnt-orange leather sofa, a patterned Persian rug, warm wooden furniture, a retro floor lamp and plants. Warm cinematic lighting, rich earthy colors, realistic textures, premium fashion campaign aesthetics. The person is casually lying on the sofa in subtle, relaxed fashion poses. CAMERA MOVEMENT (CRITICAL): each outfit begins at a cinematic diagonal side angle near the foot of the sofa. The camera smoothly moves upward into a perfectly vertical 90-degree overhead bird's-eye view, the sofa centered, and keeps a smooth clockwise rotation around its optical axis. At the end of each outfit segment the rotation accelerates into a fast, seamless spin with natural motion blur, and during the fastest part the clothing changes into the next outfit. One continuous, professionally filmed sequence. Shot 1 [0-3s], Outfit 1: a navy cap, a long blue-grey overcoat over a white shirt and a striped tie, baggy light-blue jeans, white sneakers. The person relaxes, casually raising one hand. Shot 2 [3-6s], Outfit 2: sunglasses, a navy bomber jacket over a dark shirt, wide cream trousers with a brown belt, black canvas sneakers. One hand behind the head. Shot 3 [6-9s], Outfit 3: a red cap, sunglasses, a long charcoal overcoat over a white shirt and a striped tie, wide grey trousers, white sneakers. The person adjusts the cap. Shot 4 [9-12s], Outfit 4: a black cap, a black jacket over a dark top, blue jeans, black boots. The person reads a book, then looks toward the camera. A clean overhead ending. Exactly one person, no other people, no pets. One head, two arms, two hands with five fingers each, two legs. All outfit changes happen only during the spinning transitions. The sofa, room and lighting stay the same. Photorealistic live-action, natural fabric physics, editorial fashion styling, vertical 9:16. Audio: a warm, chopped soul hip-hop instrumental loop and a quick whoosh on each fast spin. No dialogue throughout the entire video. No singing. AVOID: identity changes, outfit blending, body morphing, furniture movement, sudden camera jumps, flickering. The camera rotates around the stationary subject: the subject and the sofa never physically spin. No logos, no brand names, no text, no captions, no watermark.

2990 characters

Dressy

1 photo: you, face clearSeedance 2.512 s · 480P · 9:16152 credits
Video prompt

Create a 12-second cinematic fashion editorial video featuring the same person wearing four different outfits, based strictly on [Image 1]. The person = the person in [Image 1], in EVERY shot: copy the exact face from [Image 1]: face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles exactly as in the photo. Take only the face, the hair, the skin tone and the body shape from [Image 1], never the clothes. The same real person in all four outfits, so their friends recognize them at once. Never morph. Reproduce each outfit's clothing, colors and layering exactly; never mix clothing between outfits. SCENE: a stylish vintage living room with a burnt-orange leather sofa, a patterned Persian rug, warm wooden furniture, a retro floor lamp and plants. Warm cinematic lighting, rich earthy colors, realistic textures, premium fashion campaign aesthetics. The person is casually lying on the sofa in subtle, relaxed fashion poses. CAMERA MOVEMENT (CRITICAL): each outfit begins at a cinematic diagonal side angle near the foot of the sofa. The camera smoothly moves upward into a perfectly vertical 90-degree overhead bird's-eye view, the sofa centered, and keeps a smooth clockwise rotation around its optical axis. At the end of each outfit segment the rotation accelerates into a fast, seamless spin with natural motion blur, and during the fastest part the clothing changes into the next outfit. One continuous, professionally filmed sequence. Shot 1 [0-3s], Outfit 1: a cream lace blouse with a black ribbon bow, a cream tiered skirt to the knee, white knee socks, black shoes. The person relaxes, casually raising one hand. Shot 2 [3-6s], Outfit 2: a cropped white blazer over a white top, a white pleated skirt to the knee, white lace-up boots. One hand behind the head. Shot 3 [6-9s], Outfit 3: a fluffy cream cardigan with a black bow, a white midi skirt, white heels. The person smooths the hair and smiles. Shot 4 [9-12s], Outfit 4: a white lace blouse, a red plaid pleated skirt to the knee, white knee socks, red shoes. The person reads a book, then looks toward the camera. If [Image 1] shows a man: menswear in the same colors (a cream shirt with a black tie, a white blazer and trousers, a cream knit sweater, a red plaid overshirt), never a skirt. Nothing see-through; the camera never looks up a skirt. Exactly one person, no other people, no pets. One head, two arms, two hands with five fingers each, two legs. All outfit changes happen only during the spinning transitions. The sofa, room and lighting stay the same. Photorealistic live-action, natural fabric physics, editorial fashion styling, vertical 9:16. Audio: a warm, chopped soul hip-hop instrumental loop and a quick whoosh on each fast spin. No dialogue throughout the entire video. No singing. AVOID: identity changes, outfit blending, body morphing, furniture movement, sudden camera jumps, flickering. The camera rotates around the stationary subject: the subject and the sofa never physically spin. No logos, no brand names, no text, no captions, no watermark.

3159 characters

Pet

1 photo: your pet, face and body clearSeedance 2.512 s · 480P · 9:16152 credits
Video prompt

Create a 12-second cinematic fashion editorial video featuring the same pet wearing four different outfits, based strictly on [Image 1]. The pet = the animal in [Image 1], in EVERY shot: the same species and breed, face, eyes, ears, muzzle, fur colors, fur length and markings as in [Image 1], with the natural body volume and proportions of that animal: never flattened, squashed, stretched or merged with anything. The same pet in all four outfits, so its owner recognizes it at once. Each outfit is pet-sized and fits its body. SCENE: a stylish vintage living room with a burnt-orange leather sofa, a patterned Persian rug, warm wooden furniture, a retro floor lamp and plants. Warm cinematic lighting, rich earthy colors, realistic textures, premium fashion campaign aesthetics. The pet lies comfortably on its side or back on the sofa, relaxed, like a little fashion model. CAMERA MOVEMENT (CRITICAL): each outfit begins at a cinematic diagonal side angle near the foot of the sofa. The camera smoothly moves upward into a perfectly vertical 90-degree overhead bird's-eye view, the sofa centered, and keeps a smooth clockwise rotation around its optical axis. At the end of each outfit segment the rotation accelerates into a fast, seamless spin with natural motion blur, and during the fastest part the clothing changes into the next outfit. One continuous, professionally filmed sequence. Shot 1 [0-3s], Outfit 1: a small navy cap and a little blue-grey coat. The pet lifts one front paw. Shot 2 [3-6s], Outfit 2: tiny dark sunglasses and a little navy bomber jacket. The pet rolls a little onto its back. Shot 3 [6-9s], Outfit 3: a small red cap and a little charcoal coat. The pet yawns and stretches. Shot 4 [9-12s], Outfit 4: a small black knit beanie and a black knit sweater, an open book on the sofa beside its front paws. The pet looks up at the camera. Exactly one animal, no people, no other animals. It stays an animal with its own four legs and tail, never a person. All outfit changes happen only during the spinning transitions. The sofa, room and lighting stay the same. Photorealistic live-action, natural fabric physics, editorial fashion styling, vertical 9:16. Audio: a warm, chopped soul hip-hop instrumental loop and a quick whoosh on each fast spin. No dialogue throughout the entire video. No singing. AVOID: identity changes, outfit blending, body morphing, furniture movement, sudden camera jumps, flickering. The camera rotates around the stationary subject: the subject and the sofa never physically spin. No logos, no brand names, no text, no captions, no watermark.

2610 characters

需要多少积分?

You pay in credits. The Generate button shows this price before you spend anything. These are the options on this page.

OptionModelWhat you getCredits
StreetwearSeedance 2.512 s · 480P · 9:16152
DressySeedance 2.512 s · 480P · 9:16152
PetSeedance 2.512 s · 480P · 9:16152

One-time packs and plans are on the pricing page. A failed video gives its credits back.

沙发换装如何工作?

Every look follows the same camera move, four times in 12 seconds:

TimeWhat you see
0-3 sThe camera starts low at the foot of the sofa, rises to straight above you and turns slowly. Look 1.
3-6 sA fast spin with motion blur, and you come out of it in look 2.
6-9 sAnother spin, look 3.
9-12 sThe last spin, look 4, and the camera settles above you.

The spin is the trick: the outfit changes in the blur, so it looks like one take. That makes it an overhead outfit change, a spinning outfit change and a rotating camera outfit change at the same time: in other words, a top down outfit change with a twist. In a real sofa outfit change video, people stop the camera and change clothes for each look. The AI makes all four looks from your photo, so you lie on the sofa once and never change. Lying on sofa, outfit change after outfit change: that is the whole clip, a sofa outfit transition in one take.

The room is a warm vintage living room: a burnt-orange leather sofa, a Persian rug, wooden furniture, a retro floor lamp and plants. The sound is made with the clip: a warm, chopped soul hip-hop beat with a whoosh on each spin.

无人机旋转、街头风、正装风或宠物版:你会得到哪些造型?

点击"创建你自己的"并选择一个版本。所有四个版本都使用一张照片生成 12 秒竖屏视频。

无人机旋转逐镜头复制 AI 趋势视频本身:相同的房间、相同的升起和旋转的类似无人机的镜头及其四套街头风造型,换上你的面孔,结果配有视频自己的节拍。这是最接近趋势的复制版本。

街头风复制开启趋势的 AI 视频的四套造型:

  1. 藏青色帽子、白衬衫上穿蓝灰色长大衣配条纹领带、宽松浅蓝色牛仔裤和白色运动鞋。
  2. 太阳镜、藏青色飞行员夹克、米色宽裤配棕色腰带和黑色帆布鞋。
  3. 红色帽子、太阳镜、白衬衫配领带外穿炭灰色长大衣、宽灰色裤子。
  4. 黑色帽子、黑色夹克、蓝色牛仔裤和黑色靴子,手里拿着一本书。

正装风有四套更柔和的造型:米色蕾丝衬衫配黑色蝴蝶结和多层裙、短款白色西装套装、蓬松米色开衫配白色裙子、白色蕾丝衬衫配红色格子裙。每条裙子都到膝盖。男性照片会得到相同颜色的男装。

宠物版让你的狗或猫穿着四套小服装躺在沙发上:帽子和外套、小太阳镜和飞行员夹克、红色帽子,然后是针织毛线帽和毛衣,爪子旁边有一本书。这是沙发换装宠物版、狗狗换装 AI 视频或猫咪换装 AI 视频,取决于你的照片。

模型在你的身体上绘制衣服,所以小细节可能在每次运行时有所变化,例如领带上的条纹。每套造型的剪裁和颜色保持接近列表。

可能出现什么问题,如何解决?

问题解决方法
镜头升起但不旋转旋转是 AI 视频最难的部分。重新运行。
俯视镜头中的面部不太像你使用面部更大、更清晰、光线更好的照片,然后重新运行。
造型丢失细节(领带、书、帽子)每次运行都是新的拍摄。重新运行。
你的宠物丢失部分服装小服装在动物身上很难。重新运行。
出现第二个人或动物提示词要求一个主体。重新运行。
视频模型拒绝照片有些照片会被拒绝。积分会返还;尝试另一张照片。

结果在每次运行时都不同,所以第二次尝试通常能解决小问题。

它与吊扇趋势有何不同?

两者都是从上方拍摄的换装视频,但技巧不同。在 AI 吊扇趋势中,镜头保持静止,每次变换时吊扇叶片划过镜头。在沙发换装中没有风扇:你躺在沙发上,镜头本身旋转,所以动态模糊隐藏了变换。如果你想真实拍摄,传统方法是在沙发上方架手机支架,在造型之间剪辑,就像"4 套造型一张沙发"视频那样。如果你想要更多换装,试试电梯换装或扶梯换装。

FAQ

沙发换装趋势问题

什么是 AI 沙发换装趋势?
It is a fashion clip where you lie on a sofa and the camera spins above you; every fast spin reveals a new outfit, four in 12 seconds. The AI version showed up in late October 2026 as a sofa fashion transition template. Here it is made from one photo of you.
它与 TikTok 上的沙发换装一样吗?
它是 AI 版本。真实的沙发换装从上方拍摄,造型之间有剪辑,就像"4 套造型一张沙发"视频。AI 视频添加了旋转镜头并为你绘制造型,所以你不需要衣柜,也不需要在沙发上方架三脚架。
我如何制作沙发换装视频?
点击"创建你自己的",选择无人机旋转、街头风、正装风或宠物版,添加一张面部清晰的照片并点击生成。按钮会先显示价格。你无需编写提示词:沙发换装提示词已内置,包括镜头运动和所有四套造型。视频准备好后会在"我的视频"中等待。
有沙发换装的 CapCut 模板吗?
A Korean CapCut template called "sofa fashion transition" appeared on TikTok on 30 October 2026, and AI apps advertise an AI couch photoshoot version. If you were looking for a couch photoshoot AI clip, this template makes one. Here you do not need a sofa outfit change CapCut template: one photo and one press make the clip.
我可以用我的狗或猫制作吗?
可以。选择宠物版并添加一张你宠物的照片。它躺在沙发上,镜头在它上方旋转,每次旋转都会为它穿上新的小服装。狗和猫都可以。
我可以选择自己的造型吗?
此模板暂时不行。选择街头风或正装风可获得两组四套造型。如果你想编写自己的造型,AI 吊扇趋势有"你的造型"选项。
视频多长,有声音吗?
12 seconds, vertical 9:16, 480P, as an MP4 file. The sound is made with the clip: a soft chopped soul hip-hop beat with a whoosh on each spin. There is no copied song on it, so you can add any sound in TikTok.
哪个 AI 模型制作视频?
Smew AI 使用顶级 AI 视频模型,包括 Seedance 2.5,直接从你的照片生成视频,没有中间图片,所以面部保持接近你本人。
需要消耗积分吗?
是的。每个视频都需要积分,创建页面会在你消费前在生成按钮上显示确切价格。如果视频失败,积分会返还。
我可以使用别人的照片吗?
使用你自己的照片、你自己的宠物或已同意的成年人的照片。绝不使用名人、陌生人或儿童的照片。在 TikTok 上发布时保留 AI 标签:造型不是真实的。
Related

More like this

Template

Bby Wow Guy AI: Insert Your Face Into the Guy Dancing at a Quince Meme

Groove toward the camera at a quinceañera, take a bite of the concha, lock eyes with the lens while chewing with stuffed cheeks, then groove back into the crowd. Upload one photo to star as the bby wow guy, complete with the song.

Read more
Template

Babydoll Dance AI: Create the Trending Jamie Miller Dance With One Photo

Energetic forward bounces, swinging arms, side steps, ball changes and a high arm flourish, all timed to the music. Upload a photo of yourself or three photos of you with two friends, and generate your Babydoll dance video with the song.

Read more
Template

End of Watch Locker Scene AI: Hold Your Gear Up to the Camera From One Photo

A locker-room gear check in black tactical clothes: sidearm, light, knife, cuffs, radio, spray, then a salute. Your face, your partner, or your pet. The outfit stays the clip's, not your street clothes.

Read more
Template

AI Ceiling Fan Trend: Change Your Outfit on Every Blade Pass

The camera looks straight down through a slow ceiling fan at you on the floor. Each blade wipes across the lens, and you come back in a new outfit. Pick a pack of four looks or write your own outfits. One photo in, a vertical clip with the reel's sound out.

Read more
Template

Elevator Outfit Change AI: The Doors Close, a New Outfit Steps Out

You stand inside a steel elevator, filmed straight on. The doors slide shut, open again, and you are wearing a new outfit. Pick the women's looks (four evening gowns) or the men's looks (three outfits and a walk out), keep the reel's looks or write your own. One photo in, a vertical clip with the reel's sound out.

Read more
Template

Escalator Outfit Change AI: A New Look With Every Ride

The camera looks straight down at a long moving escalator. You sit back on the steps and ride through the frame, and every time the escalator brings you back, you wear a new outfit. Keep the reel's four looks or write your own. One photo in, a vertical clip with the reel's sound out.

Read more

制作你自己的沙发换装视频

添加一张照片并点击一次。无需编写提示词。183 积分。

No watermark. A failed video gives its credits back. Secure checkout by Stripe.