from grid.model.perception.segmentation.gsam2 import GSAM2
car = AirGenCar()

# We will be capturing an image from the AirGen simulator 
# and run model inference on it.

img =  car.getImage("front_center", "rgb").data

model = GSAM2(use_local = False)
result = model.run(rgbimage=img, prompt=<prompt>)
print(result.shape)

The GSAM2 class provides a wrapper for the GSAM2 model, which combines the power of Grounding DINO for text-based object detection with SAM2 for high-precision segmentation in RGB images.

​
class GSAM2()
​
use_local
boolean
default:
"False"

If True, inference call is run on the local VM, else offloaded onto GRID-Cortex. Defaults to False.

​
def run()
​
rgbimage
np.ndarray
required

The input RGB image of shape (M,N,3)(M, N, 3).

​
prompt
str
required

The text prompt to use for segmentation.

​
Returns
np.ndarray

The predicted segmentation mask of shape (M,N)(M, N).

from grid.model.perception.segmentation.gsam2 import GSAM2
car = AirGenCar()

# We will be capturing an image from the AirGen simulator 
# and run model inference on it.

img =  car.getImage("front_center", "rgb").data

model = GSAM2(use_local = False)
result = model.run(rgbimage=img, prompt=<prompt>)
print(result.shape)

This code is licensed under the Apache 2.0 and BSD-3 License.

Was this page helpful?

Previous
LSeg
Next