I'm trying to make a simple image search engine using Oxford Buildings Dataset First I used SIFT to extract features. Basically I iterated over every image in dataset and generate key points and descriptors for everyone of them. Then, I choose one random picture as query and used bf.match method and find the matching key points find distance between images in dataset and query.But after that -I guess this is the part that I went really wrong- I count the matching key point. The idea was which picture has the most matching key point is the picture that most similar to the query image.
matches = bf.match(query_d, desc)
# Sort the matches in the order of their distance.
matches = sorted(matches, key = lambda x:x.distance)
print matches
matches_num.append((len(matches), imageID))
My question is how can I apply any classification methods after I calculate key points?
