​
Overview

Grounding DINO (Dynamic INteractive Object detection) is a model designed to perform object detection in images based on textual descriptions. It leverages deep learning techniques to understand and locate objects described by text within an image.
​
GroundingDINO()
The GroundingDINO class encapsulates the functionality of the Grounding DINO model, providing methods for loading the model, detecting objects based on textual descriptions, and managing resources.

​
Methods

​
GroundingDINO.detect(image, classes)
  • detect(image: Union[np.ndarray, Image.Image], classes: List[str])-> Detections:
This method detects objects in the provided image based on the classes described in text. Returns a BoundingBox Detections object.

​
When to use?

One of the main benifits of Grounding DINO is that it provides a good balance, between speed and accuracy, which makes it a good choice for annotating datasets. However, it may miss out on finer details compared to slower models like OwlV2.

​
Example Usage

wget -O kites.jpg "https://github.com/overeasy-sh/overeasy/blob/main/examples/kites.jpg?raw=true"

from overeasy import *
from PIL import Image
workflow = Workflow([
BoundingBoxSelectAgent(classes=["butterfly kite"], model=GroundingDino()),
])

image = Image.open("./kites.jpg")
results, graph = workflow.execute(image)
workflow.visualize(graph)

Note how the model missed the smaller butterfly kite in the back

For more information on Grounding DINO, refer to the original paper.