StackOverflow2013

Note that there are some explanatory texts on larger screens.

plurals

PO
text
Body
copied!<p>A valid approach you can consider is the <a href="http://en.wikipedia.org/wiki/Bag-of-words_model_in_computer_vision" rel="nofollow noreferrer">Bag-of-Words model</a>.</p> <p>Basically you can do an offline computation of the target images. You can extract from those images a bunch of features in order to create a codebook with algorithms like <a href="http://en.wikipedia.org/wiki/K-means_clustering" rel="nofollow noreferrer">k-means clustering</a>. Searching for the nearest images will lead to the applications of an algorithm like <a href="http://en.wikipedia.org/wiki/Nearest_neighbor_search" rel="nofollow noreferrer">Nearest neighbor search</a> in the space of the codebook.</p> <p>For the neighbour search you can use FLANN </p> <ul> <li><a href="http://www.cs.ubc.ca/~mariusm/index.php/FLANN/FLANN" rel="nofollow noreferrer">http://www.cs.ubc.ca/~mariusm/index.php/FLANN/FLANN</a></li> <li><a href="http://opencv.willowgarage.com/documentation/cpp/flann_fast_approximate_nearest_neighbor_search.html" rel="nofollow noreferrer">http://opencv.willowgarage.com/documentation/cpp/flann_fast_approximate_nearest_neighbor_search.html</a></li> </ul> <p>Take a look also at: <a href="https://stackoverflow.com/questions/987830/visual-similarity-search-algorithm?rq=1">Visual similarity search algorithm</a></p> <p>This is only a possibility and, truth must be told, this topic is really challenging and litterature on it is really huge.</p> <p>Just some references:</p> <ul> <li><a href="http://www.cs.nott.ac.uk/~qiu/webpages/Papers/ColorPatternRecognition.pdf" rel="nofollow noreferrer">http://www.cs.nott.ac.uk/~qiu/webpages/Papers/ColorPatternRecognition.pdf</a></li> <li><a href="http://cs.brown.edu/~th/papers/Hofmann-UAI99.pdf" rel="nofollow noreferrer">http://cs.brown.edu/~th/papers/Hofmann-UAI99.pdf</a></li> <li><a href="http://www.ifp.illinois.edu/~jyang29/ScSPM.htm" rel="nofollow noreferrer">http://www.ifp.illinois.edu/~jyang29/ScSPM.htm</a></li> <li><a href="http://johnwinn.org/Publications/papers/Savarese_Winn_Criminisi_Correlatons_CVPR2006.pdf" rel="nofollow noreferrer">http://johnwinn.org/Publications/papers/Savarese_Winn_Criminisi_Correlatons_CVPR2006.pdf</a></li> <li><a href="http://www-cvr.ai.uiuc.edu/ponce_grp/publication/paper/cvpr06b.pdf" rel="nofollow noreferrer">http://www-cvr.ai.uiuc.edu/ponce_grp/publication/paper/cvpr06b.pdf</a></li> </ul>

Querying!

Guidance

An individual column

Larger individual text columns get their own page to allow for proper reading.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload