PmWiki.InternetVision History
Hide minor edits - Show changes to output
Changed lines 10-11 from:
to:
[[<<]]
Changed line 17 from:
to:
[[<<]]
Changed line 5 from:
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/poster2.png |
to:
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/ssl.png |
Changed lines 8-9 from:
to:
Humans can only ever hope to manually tag a tiny fraction of the images on the Internet. Using semi-supervised learning methods we can propagate these sparse labels to unlabeled images. The challenge is doing this when we have billions of unlabeled images.
Changed lines 3-10 from:
[[PmWiki/TinyCodes|Large-Scale Retrieval]]
Construction of compact binary descriptiors for vison
to:
Changed lines 6-9 from:
[[http://people.csail.mit.edu/torralba/tinyimages/|80 Million Tiny Images]]
One recent direction of investigation, is in a "brute force" approach to recognition, using a dataset of 80 million images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
to:
!! [[PmWiki/SSL|Semi-Supervised Learning for Gigantic Image Collections]]
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/poster2.png |
!! [[http://people.csail.mit.edu/torralba/tinyimages/|80 Million Tiny Images]]
One recent direction of investigation is in a "brute force" approach to recognition, using a dataset of 80 million images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/poster2.png |
!! [[http://people.csail.mit.edu/torralba/tinyimages/|80 Million Tiny Images]]
One recent direction of investigation is in a "brute force" approach to recognition, using a dataset of 80 million images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
Changed line 18 from:
[[http://www.cs.nyu.edu/~fergus/research/google/google.html|Improving Image Search Engines]]
to:
!! [[http://www.cs.nyu.edu/~fergus/research/google/google.html|Improving Image Search Engines]]
Changed lines 11-16 from:
Another, more recent direction of investigation, is in "brute force" approaches to recognition
Blurb on Tiny images
to:
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/poster2.png |
[[http://people.csail.mit.edu/torralba/tinyimages/|80 Million Tiny Images]]
One recent direction of investigation, is in a "brute force" approach to recognition, using a dataset of 80 million images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/airplane_query2.jpg |
[[http://people.csail.mit.edu/torralba/tinyimages/|80 Million Tiny Images]]
One recent direction of investigation, is in a "brute force" approach to recognition, using a dataset of 80 million images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
%rfloat text-align=center margin-top=5px margin-right=25px margin-bottom=5px margin-left=25px% http://cs.nyu.edu/~fergus/airplane_query2.jpg |
Added lines 1-19:
! Internet Vision
[[PmWiki/SSL|Semi-Supervised Learning for Gigantic Image Collections]]
[[PmWiki/TinyCodes|Large-Scale Retrieval]]
Construction of compact binary descriptiors for vison
[[http://people.csail.mit.edu/torralba/tinyimages/|Brute Force Recognition]]
Another, more recent direction of investigation, is in "brute force" approaches to recognition, using 100's of millions of images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
Blurb on Tiny images.
[[http://www.cs.nyu.edu/~fergus/research/google/google.html|Improving Image Search Engines]]
One possibility is to use Internet image search engines to provide a diverse set of images from which object category models may be trained. In turn, these models may be applied to collections of images from the Internet or elsewhere, enabling search by visual content (known as content-based image retrieval, CBIR), rather than the text-based searches that are currently employed.
[[PmWiki/SSL|Semi-Supervised Learning for Gigantic Image Collections]]
[[PmWiki/TinyCodes|Large-Scale Retrieval]]
Construction of compact binary descriptiors for vison
[[http://people.csail.mit.edu/torralba/tinyimages/|Brute Force Recognition]]
Another, more recent direction of investigation, is in "brute force" approaches to recognition, using 100's of millions of images gathered from the web. By using overwhelming amounts of data, very simple algorithms can perform surprisingly well.
Blurb on Tiny images.
[[http://www.cs.nyu.edu/~fergus/research/google/google.html|Improving Image Search Engines]]
One possibility is to use Internet image search engines to provide a diverse set of images from which object category models may be trained. In turn, these models may be applied to collections of images from the Internet or elsewhere, enabling search by visual content (known as content-based image retrieval, CBIR), rather than the text-based searches that are currently employed.