Image search in database

asked 2015-05-14 05:32:00 -0500

cristidadalau gravatar image

Hello,

I just start using OpenCv for a personal project and I want to do something similar with Google Image search(not so...pro :) ). I will have a database with products and their images (10k+ images), and I want to give the possibility to users to search by an uploaded image.

I need a starting point of this. How can I store informations about products images and search faster than compare all images and find the best match.

My first thought is to use FLANN, to create an index with all images and use it for searching. Is there an alternative? I'm opened to all your sugestions and advices.

Thanks

edit retag flag offensive close merge delete

Comments

1

Try Bag of Features, never tried it.

fredreload gravatar imagefredreload ( 2015-05-14 09:01:56 -0500 )edit

Bag of features or bag of words?

ines.martins gravatar imageines.martins ( 2015-11-06 08:30:42 -0500 )edit
1

Those two are the same if I remember, bag of words for images using FLANN features.

fredreload gravatar imagefredreload ( 2015-11-10 07:06:27 -0500 )edit

but bad of words is not for image classification? =x I am very confused!

ines.martins gravatar imageines.martins ( 2015-11-10 09:11:33 -0500 )edit
1

The features you get can be classified. Basically you have a vector of features and you can do bag of words with it. Since I've never worked on it, someone probably has a better answer.

fredreload gravatar imagefredreload ( 2015-11-11 08:59:00 -0500 )edit