4
$\begingroup$

I'd like to compute FeatureDistance between two images with a NetGraph:

rs = NetModel["ResNet-152 Trained on ImageNet Competition Data"];
fe = NetTake[rs, {1, -3}]

Ok, so this works:

FeatureNearest[imgs -> {"Index", "Distance"}, FeatureExtractor -> fe]

But then why doesn't this:

FeatureDistance[img1, img2, fe]

enter image description here

Is there a way to turn a NetGraph[] into a valid FeatureExtractorFunction[]?

Improve this question
$\endgroup$
1
  • $\begingroup$ Sorry why use Features stuff ? you are just computing Norm@Differences@fe@{img1,img2 } $\endgroup$
    – Fortsaint
    Commented Nov 18, 2018 at 15:27

1 Answer 1

Reset to default
3
$\begingroup$

Yes, there is a way.

net = NetChain[
   {ConvolutionLayer[32, {3, 3}], AggregationLayer[Mean]},
   "Input" -> NetEncoder[{"Image", {64, 64}, ColorSpace -> "RGB"}]
   ] // NetInitialize

enter image description here

imgs = Table[RandomImage[1, {64, 64}, ColorSpace -> "RGB"], {20}];

fe = FeatureExtraction[imgs, net]

enter image description here

FeatureDistance[imgs[[1]], imgs[[2]], fe]

0.0268319

Improve this answer
$\endgroup$
2
  • $\begingroup$ Ok great. However there's something weird going on in FeatureDistance: fe = FeatureExtraction[imgs, NetTake[rs, {1, -3}]] FeatureDistance[ExampleData[{"TestImage", "Lena"}], ExampleData[{"TestImage", "Sailboat"}], fe] the distances are only 0 or 1... unless I set DistanceFunction -> CosineDistance. Do you know what FeatureDistance does by default? $\endgroup$
    – user5601
    Commented Mar 19, 2018 at 13:57
  • $\begingroup$ @user5601 EuclideanDistance. i.sstatic.net/ky1wD.png $\endgroup$
    – Alexey Golyshev
    Commented Mar 20, 2018 at 8:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.