The Copyleaks Object Detection API locates specific objects within an image and returns a pixel-level segmentation mask for each one it finds. You supply the list of objects to look for, and the API returns a Run-Length Encoded (RLE) mask per detected object in a single synchronous call. Object detection performs segmentation only - it does not classify whether an image is AI-generated. For AI-generated content detection, use the AI Image Detection API.

​
Capabilities

Targeted Object Segmentation

Provide up to three objects to locate per image and receive a precise per-object segmentation mask for each one found.

Pixel-Level Masks

Each detected object comes with a Run-Length Encoded (RLE) mask you can decode to highlight, crop, or blur the object’s region.

Synchronous Results

Submit an image and receive the detected objects in the same API call - no webhook required.

Supported Formats

Process JPEG, PNG, WebP, BMP, and HEIC/HEIF images up to 32MB and up to 27 megapixels in resolution.

​
How it works

  1. Submit an image via multipart/form-data along with an objectDetection list of the objects to locate.
  2. The API segments the image and returns a detectedObjects array - one entry per object found, each with its RLE mask.
  3. Decode the masks using the image dimensions returned in imageInfo to visualize or act on each object’s region.
Objects that are requested but not found are omitted from the response, so look entries up by their object name rather than by position.

​
Integrated Ecosystem

AI Image Detection

Detect whether an image is AI-generated and gain insights into its origin.

Image Plagiarism Detection

Search the web for unauthorized copies and usages of your images.

​
Next Steps

Start Building

Get started with the Object Detection API and integrate it into your workflows.

Endpoint Documentation

Complete technical specifications for the Object Detection endpoint.