"Detect Object Using deep learning" appears not to honor the specified extent

[VincenzoDeriu]

Hi,

I'm trying to run the "Detect Object Using Deep Learning" tool on ArcGIS Pro, giving the World Imagery basemap as input. To limit the area of the detection, I also set the "current display" extent as environment variable, but I think the tool is not honoring it, since it is stuck at '0%' for many hours.

Did someone else experienced the same issue and has some suggestion?

Thanks

[JeffreySwain]

Hi Vincenzo,

I would consider a different input layer for your processing as the World Imagery basemap is probably not the best choice in this situation. Depending on where you are looking to analyze, I would strongly consider downloading the images and using them as the input. 

Here is a blog that shows some of the sources you can use to download your own imagery.

 Imagery sources to power your analysis 

[BobBooth1]

You may find these tutorials helpful:

https://learn.arcgis.com/en/paths/try-deep-learning-in-arcgis/

[PavanYadav]

@VincenzoDeriu try setting a cell size as well. you can look up what cell size (e.g. 0.3) your model is expecting.