Attach URLs

You can import items by linking them from external storages. The linked items are displayed in SuperAnnotate, but they aren't stored on our local servers.

Attach images URLs

You can attach image URLs to Pixel Projects and Vector Projects only.

Follow the steps below to learn how to attach image URLs. We've also prepared recipes to walk you through attaching image URLs from different types of cloud storage.

Step 1: Create a CSV file containing the image URLs

To attach image URLs, first create a CSV file that contains the URLs. Use this template CSV file as a reference.

Key

Required/Optional

Value

url

Required

Image URL

name

Optional

Image name

📘

If the image name field is empty, a name ID will be generated based on the row ID and a random UUID.

Step 2: Attach image URLs via Python SDK

Link the images to your project with this SDK command.

superannotatecli attach-image-urls --project <project_name/folder_name> --attachments <csv_path> [--annotation_status <annotation_status>]
sa.attach_items(project, attachments, annotation_status = “NotStarted”)

External storage project

After attaching image URLs, the projects will be considered as an external storage project.

In an external storage project, you can't:

  • Run smart segmentation and smart prediction.
  • Run model training.

Limitations

There is no image size or resolution limit. Note that large or high-resolution images may impact the performance of the platform.

Recipes

Attach video URLs

You can attach video URLs to Video Projects only.

Step 1: Create a CSV file containing the video URLs

To attach video URLs, first create a CSV file that contains the URLs. Use this template CSV file as a reference.

Key

Required/Optional

Value

url

Required

Video URL

name

Optional

Video name

📘

If the video name field is empty, a name ID will be generated based on the row ID and a random UUID.

Step 2: Attach video URLs via Python SDK

Link the videos to your project with this SDK command.

sa.attach_items(
    project = "Project Name", 
    attachments = "./videos.csv",
    annotation_status = "QualityCheck")
superannotatecli attach-video-urls --project “Project Name/Folder Name” --attachments “./videos.csv” --annotation_status “QualityCheck”

Supported video file formats

SuperAnnotate supports the following video file formats that HTML5 supports: OGG, WEBM, and MP4. Learn more about browser compatibility.

Attach document URLs

Step 1: Enable CORS in your storage

Make sure to enable cross-origin resource sharing (CORS) for the storages you will be using.

Amazon S3 bucket

  1. In your bucket, go to Permissions.
  2. Turn off Block all public access.
  3. Insert this configuration in the CORS section.
[
    {
        "AllowedHeaders": [],
        "AllowedMethods": [
            "GET"
        ],
        "AllowedOrigins": [
            "*"
        ],
        "ExposeHeaders": []
    }
]
  1. Click Save changes.

Step 2: Create a CSV file containing the document URLs

To attach document URLs, create a CSV file that contains the URLs. Use this template CSV file as a reference.

Key

Required/Optional

Value

url

Required

Document URL

name

Optional

Document name

📘

If the document name field is empty, a name ID will be generated based on the row ID and a random UUID.

Step 3: Attach the document URLs via Python SDK

Link your documents to the project with this SDK command.

sa.attach_items(
    project = "Project Name", 
    attachments = "./documents.csv",
    annotation_status = "QualityCheck")
superannotatecli attach-document-urls --project “Project Name/Folder Name” --attachments “./documents.csv” --annotation_status “QualityCheck”

Did this page help you?