DepthAnything

The DepthAnything class implements a wrapper for the DepthAnything model, which estimates depth maps from RGB images. The model configurations are defined based on different encoder types.

class DepthAnything()

use_local · boolean · default: "True"

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

This model is currently not available via Cortex.

def run()

rgbimage · np.ndarray · required

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

Returns · np.ndarray

The predicted output of shape $(M,N)$.

Inference Call

from grid.model.perception.depth.depth_anything import DepthAnything
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 = DepthAnything(use_local = True)
result = model.run(rgbimage=img)
print(result.shape)

License

This code is licensed under the Apache 2.0 License.

Source

https://github.com/LiheYoung/Depth-Anything