Welcome! Please contribute your ideas for what challenges we might aspire to solve, changes in our community that can improve machine learning impact, and examples of machine learning projects that have had tangible impact.
Lacking context for this site? Read the original paper: Machine Learning that Matters (ICML 2012). You can also review the slides.
How to apply applying KNN classifier to different sized images represented by Bag Of Features
  • Suppose I have an image represented by a histogram created by Bag of Words approach and there are other training instances also have same number of bins for their histogram representation but train images are sharing same size as NxN. My new novel image has the size bigger than the train images thus although its histogram has the same number of bins its frequency of words on the histogram is larger, because of its size is larger. For applying an KNN classifier, I think that I need to have some kind of normalization on the histograms to avoid the effect of this size difference between images. What kind of normalization or any other process should I consider?
  • 1 Comment sorted by
  • Have you tried normalizing your bin values by the total number of pixels in the image?  That would give you a probability distribution, and you can compare them directly.


To post or add a comment, please sign in or register.

Tip: click the star icon to bookmark (follow) a discussion. You will receive email notifications of subsequent activity.
If search doesn't work, try putting a + in front of your search term.