IS = { zkontrolovano 16 Dec 2007 },
  UPDATE  = { 2007-12-12 },
   author =      {Krystian Mikolajczyk and Ji{\v r}{\' i} Matas},
   title =       {Improving SIFT for Fast Tree Matching by Optimal Linear Projection},
   year =        {2007},
   pages =       {8},
   booktitle =   {ICCV 2007: Proceedings of Eleventh IEEE International
                  Conference on Computer Vision},
   editor =      {Dimitris Metaxas and Baba Vemuri and Amnon Shashua and Harry Shum},
   publisher =   {IEEE Computer Society Press},
   address =     {Los Alamitos, USA},
   isbn =        {978-1-4244-1631-8},
   book_pages =  {2240},
   month =       {October},
   day =         {14-20},
   venue =       {Rio de Janeiro, Brazil},
   organization ={IEEE Computer Society},
   annote = {We propose to transform an image descriptor so that
     nearest neighbor (NN) search for correspondences becomes the
     optimal matching strategy under the assumption that inter-image
     deviations of corresponding descriptors have Gaussian
     distribution. The Euclidean NN in the transformed domain
     corresponds to the NN according to a truncated Mahalanobis metric
     in the original descriptor space. We provide theoretical
     justification for the proposed approach and show experimentally
     that the transformation allows a significant dimensionality
     reduction and improves matching performance of a state-of-the art
     SIFT descriptor. We observe consistent improvement in
     precision-recall and speed of fast matching in tree structures at
     the expense of little overhead for projecting the descriptors
     into transformed space. In the context of SIFT vs. transformed
     MSIFT comparison, tree search structures are evaluated according
     to different criteria and query types. All search tree
     experiments confirm that transformed M-SIFT performs better than
     the original SIFT.},
   keywords =    {SIFT, retrieval, metric treesr},
   authorship =  {50-50},
   note =        {CDROM},
   project =     {GACR 102/07/1317},
   psurl       = {pdf [259k] },
   acceptance_ratio = { 23.5% },