Object detection

Object Detection models allow users to identify objects of certain defined classes. These models receive an image as input and output the images with bounding boxes and labels on detected objects.

For more details about the object-detection task, check out its dedicated page! You will find examples and related materials.

Recommended models

facebook/detr-resnet-50: Solid object detection model pre-trained on the COCO 2017 dataset.

Explore all available models and find the one that suits you best here.

Using the API

Language

Client

Provider

Settings

from huggingface_hub import InferenceClient

client = InferenceClient(
    provider="hf-inference",
    api_key="hf_xxxxxxxxxxxxxxxxxxxxxxxx",
)

output = client.object_detection("cats.jpg", model="facebook/detr-resnet-50")

API specification

Request

Headers
authorization	string	Authentication header in the form `'Bearer: hf_**'` when `hf_**` is a personal user access token with “Inference Providers” permission. You can generate one from your settings page.

Payload
inputs*	string	The input image data as a base64-encoded string. If no `parameters` are provided, you can also provide the image data as a raw bytes payload.
parameters	object
threshold	number	The probability necessary to make a prediction.

Response

Body
(array)	object[]	Output is an array of objects.
label	string	The predicted label for the bounding box.
score	number	The associated score / probability.
box	object
xmin	integer	The x-coordinate of the top-left corner of the bounding box.
xmax	integer	The x-coordinate of the bottom-right corner of the bounding box.
ymin	integer	The y-coordinate of the top-left corner of the bounding box.
ymax	integer	The y-coordinate of the bottom-right corner of the bounding box.