Workspace Image Query
The Workspace Image Search API provides a powerful and flexible way to search and list images within your workspace. It allows users to filter, sort, and perform semantic searches using a query string that is similar to the in-app search functionality.
Using the continuation token returned by the API it is possible to make follow up request to page through all images that match a certain query without (e.g. for purposes of listing all images in a workspace or specific projects)
This documentation details the usage of the API endpoint.
Endpoint
POST https://api.roboflow.com/{WORKSPACE}/search/v1
This endpoint supports searching for images based on various criteria specified in the request.
Authentication
To access the API, you need to include your API key in the request. The API key should be passed as a query parameter.
Example: ?api_key={YOUR_API_KEY}
Request Format
Headers
Content-Type: application/json
Body Parameters
query
: A string to filter, sort, or perform a semantic search. For Example: "nightime project:{my-project-url}" will filter for images in themy-project-url
project and apply a semantic sort for images that matchnighttime
.pageSize
: Number of results to return per page (default: 50).fields
: List of fields to include in the response. Possible values are "tags", "width", "height", "filename", "aspectRatio", "split".
Example Request
Response Format
The response is a JSON object containing the following fields:
results
: An array of image objects.total
: Total number of images found.continuationToken
: A token for pagination.
Image Object Fields
depending on fields requested in the fields
parameter:
id
: Unique identifier of the image.projectData
: Object containing project-specific data.split
: Indicates the dataset split (e.g., "test").inDataset
: Boolean indicating if the image is in the dataset.
tags
: Array of tags associated with the image.width
: Width of the image in pixels.height
: Height of the image in pixels.filename
: Name of the image file.aspectRatio
: Aspect ratio of the image.
Pagination and continuationToken
continuationToken
The continuationToken
returned in the response is crucial for pagination. It allows users to page through all the matching search results efficiently. When the continuationToken
is included in a subsequent request, it fetches the next set of results.
To use the continuationToken
, include it in the request body of your next API call. This will retrieve the next page of results based on your original query.
Project and Dataset Filters
In addition to all the filters available in the in app dataset search, the workspace wide search also support filters for project
and dataset
. For example you can use a query like project:foo project:bar
to find all images that are in both projects foo
and bar
.
A query like dataset:foo
will return images in project foo
that have been labeled and are added to the dataset in the project.
Last updated