API Overview

Clipping Magic offers a REST-oriented Web API for automating the uploading of images and downloading of results. You can integrate the Clipping Magic Editor into your website with ClippingMagic.js to let your users remove backgrounds right on your website.

The API is especially useful for websites needing to remove backgrounds from user-provided images, and for eCommerce websites with a staff of graphic designers removing image backgrounds on a flow of images.

The API offers the unique ability to do both review and touch up in a single flow. That means a single step to have every image reviewed and ready for use. In contrast, AI-only APIs force you to have a separate review process with some fallback mechanism for failing images.

To use the API you will need an API Key and some web programming skills.

Theory Of Operation

Auto-Clip Operation

You can leverage Clipping Magic's Auto-Clip feature to clip images fully automatically and without any human operator in the loop. You simply upload an image and get the result as the response.

While Auto-Clip works excellently on a wide range of images, we always recommend validating that your particular image flow is handled to your satisfaction.

Hair is not yet handled fully automatically. If your image flow contains pictures with hair in them we recommend using a human-in-the-loop workflow for the time being.

Internal Staff Removes Backgrounds

The typical scenario is that you have a long list of images that need to be processed, and some staff to process them. The steps you would normally take are:

  1. Upload a batch of images one at a time to ClippingMagic.com via the backend API. Get and store the respective image ids and secrets.

  2. In your internal system, allocate images to workers in suitable batches. Then let each worker remove the backgrounds from a batch with ClippingMagic.js, using the Edit Multiple Images In Sequence workflow with the corresponding image ids and secrets from step 1.

    Typical batch sizes vary between 10 and 100 depending on volume and the particulars of your images.

  3. Upon receiving the result-generated callbacks from ClippingMagic.js notify your backend to download the result for that image.

  4. Rinse and repeat.

End User Removes Backgrounds

The typical scenario is that an end-user uploads an image to your site, and you want to enable the user to remove the background from that image as part of your site image workflow. The steps you would normally take are:

  1. Your user uploads an image to your website.

  2. Your backend server passes the image to ClippingMagic.com via the backend API and gets an image id and secret back.

    Your web client cannot upload the image directly to ClippingMagic.com for security reasons.

  3. Your backend server passes the image id and secret to your web client. If you're using AJAX this information can be included in the response to the image upload. Otherwise it can be embedded in the webpage you render in response to the user's image upload.

  4. On your site, let the end-user remove the background with ClippingMagic.js, by having your javascript code invoke the Edit A Single Image workflow with the image id and secret from the previous step.

  5. Upon receiving the result-generated callback from ClippingMagic.js your javascript code notifies your backend to download the result for that image.

  6. Your backend server downloads the result and continues with the subsequent steps in your site-specific workflow.

Evaluating The API

It is free to create an API Key and to submit test images for processing. Test images work just like regular images, but their results have a watermark added to them. This allows you to fully evaluate the API for free and without a subscription.

Once you have evaluated the API you can sign up for an API subscription.

Details

  • You can use the API in test mode without signing up for a subscription, but in order to use it for real you will need to sign up for an API subscription.
  • Sorry, regular website subscriptions do not come with API access.
  • You can view and edit all images uploaded through the API on the main Clipping Magic website (when logged in as the API user) and vice versa. This can be very useful when debugging, or if you need to review the API image flow.
  • Your http client library must support Server Name Indication (SNI) to successfully make requests. If you're getting weird handshake errors, this is most likely it.

Feedback

Is something missing or insufficiently documented? Does your implementation require additional features?

Be sure to let us know and we'll try to sort it out!