Upload Image
Bring your own picture into a workflow. Upload a file or paste a link, crop it on the way in, and send it to every node that takes an image.
The Upload Image node brings a picture of your own into a workflow. You upload a file or paste a link to an image, and the node passes that picture to every node you connect it to. Use it for reference photos, product shots, logos and any still you want an AI node to work from.
When to use it
- You have a photo of a person or a product and want Generate Image to keep it recognizable. Wire the photo into the References input.
- You want to animate a still. Wire it into Generate Video as the first frame.
- You want to edit a photo. Send it to Modify Image, Upscale Image or Remove Background.
- You want a logo or a watermark on your results. Wire the file into a layer of Image Overlay.
Quick start
Add the node
Press Tab on the canvas and choose Image › Add Your Own › Upload Image.
Add a picture
Click Choose Image, or drop an image file on the node. To use a picture that is already online, paste a direct link to the image into the field on the node instead.
Crop or convert it
When you choose a file, the Adjust Image step opens before the upload. Crop the picture with an aspect-ratio preset, such as Square or Widescreen, or with Freeform. Under Advanced, you can convert the file to PNG, JPEG or WebP.
Connect it
Drag from the node's Image output to any node that takes an image. Every connected node uses the picture when it runs.
Output
Upload Image is a source node: it needs nothing wired into it. Its output, Image, is the link to your picture. Every node that accepts an image can take it, and one Upload Image node can feed any number of nodes at once.
Settings
| Setting | What it does |
|---|---|
| Choose Image | Opens your files. You can also drop a file on the node. |
| Image link | A direct link to an image, pasted into the field on the node instead of a file. The link must be public. |
| Adjust Image | Opens when you choose a file. Crop with an aspect-ratio preset or Freeform, and convert to PNG, JPEG or WebP under Advanced. The crop runs in your browser before the upload. |
| Label | The node's name on the canvas. The label also becomes the name you use to mention the picture in a prompt. |
The node accepts common image formats, such as PNG, JPEG and WebP.
Mention the picture in a prompt
When you wire an Upload Image node into Generate Image, you can refer to the picture by name inside the prompt. The name comes from the node's label: a node labeled "Town" becomes @town:1.
| You write | The model receives |
|---|---|
@town:1 | The picture itself, placed where you typed the mention |
@town:1:background | Only the background of the picture |
@town:1:signage | Any single word you choose, used as written |
After the second colon, you can add one of these roles: object, person, face, clothes, background, style, pose or texture. Any single word of your own works too. Mentioning is optional: an unnamed upload is still sent to the model as a reference.
A label that starts with a digit, such as "3D Render", cannot form a mention. Rename the node if you want to mention it. When a character or a location has the same name, the character or location wins. Read Reference roles for every role and for the identity lock.
Tips
- Start from a good picture. A sharp, well-lit, high-resolution image gives better results in every AI node that uses it.
- Crop for the target. If the image will become the first frame of a vertical video, crop it to a vertical preset in the Adjust Image step.
- Use public links. A link that needs a login or expires does not work. Upload the file instead.
- Videos from social platforms need another node. To bring in a clip from a social platform, use Video URL, which handles platform downloads.
Frequently asked questions
Related
Generate Image
Modify Image
Generate Video
Reference roles
Assets and media
Last updated on
Image nodes
Every image node in Nodaro, from uploads and text-to-image to editing, masks, collages, overlays, reference boards and image analysis.
Generate Image
Create an image from a text prompt with one of Nodaro's image models, keep characters consistent with references, and edit results in place with a mask.