image-processingmachine-learningcomputer-visioncontent-based-retrieval

How to measure distance between Fisher Vector for Image Retrieval?


I've read something about Fisher Vector and I'm still in the learning process. It's a better representation than the classic BoF representation, exploiting GMM (or k-means, even if that's usually referred as VLAD).

However, I've seen that usually they are used for classification problem, for example with SVM.

But what about Image Retrieval? I've seen that they have been used for image retrieval too (here), but I don't understand one point: given two FV representing 2 images, how do we compute their distances and so "how similar the two images are?"

Is it reasonable to use them in such a context?


Solution

  • As seen in the two papers below, Euclidean distance seems to be the popular choice. There are also references to using dot-product as a similarity measure; cosine similarity (closely related) is a generally popular metric for ML similarity.

    http://link.springer.com/article/10.1007/s11263-013-0636-x

    http://www.robots.ox.ac.uk/~vgg/publications/2013/Simonyan13/simonyan13.pdf

    Is this enough to let you choose something that meets your needs?