Wolfram Function Repository

Instant-use add-on functions for the Wolfram Language

Function Repository Resource:

 ImageGeoLocation

Source Notebook

Find the location at which a picture was taken

Contributed by: Giulio Alessandrini

ResourceFunction["ImageGeoLocation"][img]

returns the estimated GeoLocation[…] at which img was taken.

ResourceFunction["ImageGeoLocation"][img,n]

returns the first n probable locations.

ResourceFunction["ImageGeoLocation"][img,n,"Probabilities"]

returns the locations together with their probabilities.

ResourceFunction["ImageGeoLocation"][{img1,img2,…},…]

estimates multiple locations at once.

Details

ImageGeoLocation uses the neural network ResNet-101 to determine probable locations.

Examples

Basic Examples (2) 

Find the location at which a picture was taken:

In[1]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/a6ff96a5-1885-41b9-b53b-e4bb6449fbed"]
Out[1]=

Find multiple location candidates:

In[2]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/91d1ffc2-81b1-488c-878f-020af2122daf"]
Out[2]=

Map them:

In[3]:=
GeoGraphics[GeoMarker[positions]]
Out[3]=

Scope (5) 

Predict one location:

In[4]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/70773b9d-65ab-4ab3-9df2-3fe0599970fc"]
Out[4]=

Return the first n most probable locations:

In[5]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/64356d01-e2d4-463a-8a8e-1b41985c55e5"]
Out[5]=

Predict locations for multiple images:

In[6]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/ca5d879f-957a-4076-a0df-15878afb94d7"]
Out[6]=

Return the first n most probable locations for multiple images:

In[7]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/7af117de-85ca-4e4c-857d-ab363d3d7406"]
Out[7]=

Return locations together with their probabilities:

In[8]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/bdba0f70-0b53-470c-a490-f7b5314ceb0b"]
Out[8]=
In[9]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/1dd85709-bc4b-430f-8294-e61ceb218231"]
Out[9]=

Applications (2) 

In places with high population density, very fine-grained predictions are possible. Consider the following four landmarks in Paris:

In[10]:=
landmarks = EntityValue[{Entity["Building", "EiffelTower::5h9w8"], Entity["Building", "TheLouvre::vqy3g"], Entity["Building", "NotreDameCathedral::95fcw"], Entity["Building", "ArcDeTriomphe::92x88"]}, "Image", "EntityAssociation"]
Out[10]=

Predict the location of the four landmarks:

In[11]:=
locations = ResourceFunction["ImageGeoLocation"] /@ landmarks
Out[11]=

Plot the locations on the map:

In[12]:=
GeoGraphics[GeoMarker /@ locations]
Out[12]=

Compare to the locations from Wolfram|Alpha:

In[13]:=
KeyMap[GeoPosition, locations]
Out[13]=

Find the most probable locations for an image of white cliffs:

In[14]:=
(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/77a42f5c-e513-4c14-bbef-f8f0737b48e8"]
Out[14]=

Plot them in a bubble chart where the radius is given by the probability value:

In[15]:=
GeoBubbleChart[locations, GeoRange -> Quantity[200, "Kilometers"]]
Out[15]=

Version History

  • 1.0.0 – 25 January 2023

Related Resources

License Information