Detecting and Sketching the Common
Shai Bagon,
Or Brostovsky,
Meirav Galun,
Michal Irani
This webpage presents the paper "Detecting and Sketching the Common" (CVPR 2010).
Paper [PDF] [bibtex]
Presentation [PPTX]
Sketching code.
Conference spotlight presentation.
Abstract
Given very few images containing a common object of interest under
severe variations in appearance, we detect the common object and
provide a compact visual representation of that object, depicted
by a binary sketch. Our algorithm is composed of two stages:
(i) Detect a mutually common (yet non-trivial) ensemble of
`self-similarity descriptors' shared by all the input images.
(ii) Having found such a mutually common ensemble, `invert' it to
generate a compact sketch which best represents this ensemble.
This provides a simple and compact visual representation of the
common object, while eliminating the background clutter of the
query images. It can be obtained from very few query images.
Such clean sketches may be useful for detection, retrieval,
recognition, co-segmentation, and for artistic graphical purposes.
Results
A sample of our qualitative results on a few challenging sets.
More examples can be found in the paper
and in the PPTX presentation.
