Bag of Visual Words (BoVW) implementation in python based on OpenCV Computer Vision with Python.
Bag of Visual Word is simple technique to represent images by using only their descriptors. BoVW can be used as a dimensionality reduction procedure on your image dataset before a classification task.
You can use sift, surf or orb as feature extractor in this BoVW. For the last two you have to compile the opencv library from source. Also the last two feature extractors are not for commercial usage as you can find in opencv documentation.
git clone git@github.com:theopsall/BoVW.gitfrom BoVW import bovw