SURF for image retrieval
3 views (last 30 days)
Show older comments
Dear Team,
I've few doubts in implementation of SURF? In most of the web tutorials i've read SURF seem to be focussed on a simple comparison between two images. Instead of determining how similar two images are, would it be practical to use SURF to find the N closest matching images out of a collection of thousands of images?
For example, would it be practical to use SURF to generate keypoints for a batch of images, store the keypoints in a database, and then find the ones that have the shortest Euclidean distance to the keypoints generated for a "query" image?
Looking forward for your valuable suggestions and guidance on how to proceed if this could be done?
Malini
0 Comments
Accepted Answer
Image Analyst
on 29 Nov 2013
You can use SIFT and SURF for "scene matching" and Content Based Image Retrieval. I've seen some impressive results. Though with SIFT, since it's patented, that can be a problem. Many people, including the Mathworks in their Computer Vision System Toolbox, use SURF which is about as good as SIFT and much faster. With CBIR, there are a lot of different features and algorithms that can be used to suggest similar images. It's too involved to get into here, plus it's a rapidly evolving field of study. You can get MATLAB SIFT code at http://www.robots.ox.ac.uk/~vedaldi/code/sift.html, or http://www.vlfeat.org/
4 Comments
More Answers (0)
See Also
Categories
Find more on Computer Vision Toolbox in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!