IS = { zkontrolovano 15 Jan 2013 },
  UPDATE  = { 2012-12-27 },
  editor     = {Fitzgibbon, Andrew and Lazebnik, Svetlana and 
                Perona, Pietro and Sato, Yoichi and Schmid, Cordelia},
  booktitle  = {Computer Vision - ECCV 2012 - Part II},
  publisher  = {Springer},
  address    = {Heidelberg, Germany},
  series     = {Lecture Notes in Computer Science},
  volume     = {7573},
  year       = {2012},
  isbn       = {978-3-642-33708-6},
  author     = {J{\' e}gou, Herv{\' e} and Chum, Ond{\v r}ej},
  title      = {Negative Evidences and Co-occurences in Image Retrieval: 
                The Benefit of PCA and Whitening},
  pages      = {774--787},
  book_pages = {889},
  month      = {October},
  day        = {7-13},
  venue      = {Florence, Italy},
  annote = {The paper addresses large scale image retrieval with short
    vector representations. We study dimensionality reduction by
    Principal Component Analysis (PCA) and propose improvements to its
    different phases.We show and explicitly exploit relations between
    i) mean subtraction and the negative evidence, i.e., a visual word
    that is mutually missing in two descriptions being compared, and
    ii) the axis de-correlation and the co-occurrences
    phenomenon. Finally, we propose an effective way to alleviate the
    quantization artifacts through a joint dimensionality reduction of
    multiple vocabularies. The proposed techniques are simple, yet
    significantly and consistently improve over the state of the art
    on compact image representations. Complementary experiments in
    image classification show that the methods are generally
  keywords =   {image retrieval, short codes, PCA},
  prestige =   {important},
  authorship = {50-50},
  note =       {CD-ROM},
  project =    {GACR P103/12/2310, Quaero project by OSEO},
  psurl =      {http://cmp.felk.cvut.cz/~chum/papers/Jegou-ECCV12.pdf},