Clothing photo to model turnaround video

What it is

This n8n workflow takes a photo of a garment, dresses a model in it in a plain studio and makes a 10-second video of that model turning around. Eighteen nodes run in one line. Three sticky notes name the stages. A form takes the photo and Dropbox keeps it behind a public link. A Flux model on Runware makes the studio shot from it, and Kling, also on Runware, animates the turn. No language model is involved. Both prompts are fixed text in the file.

1. A form takes one clothing photo
2. Dropbox keeps it behind a public link
3. Flux on Runware dresses a model in it
4. Kling on Runware films a slow turn
5. The video’s address lands in the run

Who it suits

Clothing shops with flat-lay or mannequin photos that want an on-model clip for each product page. The prompts choose the model’s gender from the garment and set a plain studio. Per run you control only the photo. Skip it for anything about fit. The model is generated, so the drape and the size on screen are a guess, not a fitting.

Where to use it

Any n8n that can reach Dropbox and Runware will run it. The finished video stays at an address on Runware.

• n8n Cloud: import the file into a new workflow there
• Your own n8n server: the same file, the same steps
• Product pages and ads: download the clip from its address and upload it where the product is sold
• A drive or a sheet: add a node after Edit Fields1 to keep the address or save the file

Setup

Get it saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Its author puts setup at about 9.5 minutes.

1. In n8n, create a workflow, open its menu, choose Import from File and pick workflow.json.
2. In Dropbox’s App Console, give your app the files.content.write, sharing.read and sharing.write permissions, then generate an access token.
3. Add a Dropbox credential with that token and select it in Upload a file.
4. In HTTP Request1 and HTTP Request6, replace <YOUR_DROPBOX_ACCESS_TOKEN> after Bearer with the same token. The brackets go too.
5. In Flux Image, kling image to video post and kling imag to video get status, replace <YOUR_RUNWARE_API_KEY> the same way with your Runware key.
6. Open On form submission and rewrite its description, which still promises a CGI ad video.
7. Save, send a test photo through the form’s test address, then press Publish.

A pasted Dropbox token is short-lived now and stops working after a while. To stop pasting new ones, give Upload a file a Dropbox OAuth2 credential. Then point both HTTP Request nodes’ Authentication at it and delete their Authorization header.

How to use it

Open the form from On form submission, attach one photo of the garment in upload product image and send it. The form confirms at once. Behind it the run needs at least three minutes and forty seconds, the two waits added up. Open it under Executions and read finalVideoURL in Edit Fields1.

Example

A flat-lay photo of a navy denim jacket goes through the form. Runware gets this body from Flux Image, its fixed prompt cut short here:

[
  {
    "taskType": "imageInference",
    "positivePrompt": "Place a human model wearing this clothing product. …",
    "referenceImages": ["<the photo’s Dropbox address>"],
    "model": "bfl:3@1",
    "height": 1024,
    "width": 1024,
    "outputType": "URL",
    "outputFormat": "PNG"
  }
]

The rest of that prompt asks for a model whose gender suits the jacket, a plain studio, and the jacket left exactly as photographed. The square studio shot then opens a 10-second, 1280 by 720 video in which the model turns to show the jacket from every side.

For newcomers and experts

If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. A credential is a key n8n keeps for you and hands to the node you pick it in. This file also puts tokens straight into five headers, so you paste each one by hand. Send a photo from the form’s test address first. Then open each node to see what it got.

For experts. Runware does both steps through one endpoint. Flux Image sends your photo to bfl:3@1 as a reference image. Then kling image to video post starts klingai:5@1 on the studio shot asynchronously, so Split Out keeps the task’s id and a later getResponse call fetches the video.

The studio shot is square. The video is not, so Kling starts from a square picture. Change the first body’s width and height if the turn looks cropped.

Good to know

The garment can drift, and the run checks on the video once.

• The garment: both prompts tell the models to keep it as it is, but a print, a logo or a seam can still change. Compare the clip with your photo before it goes on a product page
• One status check: a video still rendering after the wait leaves finalVideoURL empty, so lengthen Wait or loop back to it until a videoURL arrives
• The original copy: its Upload a file cleaned names with /s+/g, which hits the letter s. This copy has the backslash back
• Billing: Runware bills each run twice, for the studio shot and for the video
• Your photo: it stays at the top of your Dropbox with a public link, and it also goes to Runware
• Open form: once the workflow is published, anyone with the form’s address can start a run on your key, so turn on the form trigger’s basic authentication or keep the address to yourself
• Credentials: the JSON holds none, and nothing from its author’s account