How to verify if two images have the same object/content?
Hello all.
I am truly new to computer vision field, but it is fascinating me! I have now a challenge in my hands and I am seeking for mentors/advisers to give me some follow up.
My project is: From a picture of a video-game cover, search that pic through a video-game cover database, and if there is a very good match the app will return a string with the name of the videogame and the platform.
Problem example: 1 -Take a photo of a cover similar to this: http://i.imgur.com/gpZMbRm.jpg 2 - Cover match to this in the database: http://i.imgur.com/WXhPwf8.png 3 - App gives string: "Fifa 12 Playstation 2"
In my preliminary research I found that I should save in my cover database, the name of the game, the platform, the URL for the cover, and the image features (keypoints and descriptors).
I am using SURF features detector/extractor. My first trials output is something like this: http://i.imgur.com/Rj9lcBr.png .
There are some concepts that I am still confused about... I am not looking for similarity right? I just need to observe if there are some sort of good keypoints-paired/matched, right? Because in the examples images mentioned before I get "img1 - 1087 features (query image), img2 - 1755 features - 30 % - SIMILARITY - 321/333 inliers/matched" What are the inliers? My calculation of similarity seems wrong to me... I would say that these two images are like 70 % look alike...
ps: I am using Python so...
Thanks for your time and help. Sorry If I could not explain any better my problems/concerns.
If you are using Python, you should read these tutorials: http://docs.opencv.org/master/db/d27/tutorial_py_table_of_contents_feature2d.html#gsc.tab=0.
@Eduardo I have already done some of those tutorials (very good indeed) and I am in some advanced stage now. I am just confused at how to tell or set a threshold that above for example 100 kp_pairs/matches is a high probability of be the same game between pics
I have done similar kind of work with cascades. You can try this using lbph algorithm. You can get the confidence value which will clearly state how close the image is to the original one. Hope this helps.