Perceptual Feature Detection

@InProceedings\{MOV2009,
  author       = "Naila Murray and Xavier Otazu and Maria Vanrell",
  title        = "Perceptual Feature Detection",
  booktitle    = "New Trends and Challenges in Computer Vision: Progress of Research and Development",
  year         = "2009",
  abstract     = "Currently there exists no application-independent
or general theory of feature detection. In this work,
a brightness induction wavelet model (BIWaM) is
extended with the long-term aim of developing a
principled model for generic local feature detection.
This detector, the Feature Induction Wavelet
Model (FIWaM), uses the same \“featureness\” measure
for a range of local features such as blobs, bars
and corners. FIWaM is a wavelet-based computational
model that attempts to use the perceptual
processes involved in visual brightness induction
to enhance and detect these features. The model
uses two center-surround mechanisms in sequence
to detect features - a Gabor-like mother wavelet
followed by an explicitly-defined center-surround
region mechanism. These center-surround regions
are feature-specific and introduce the only variation
in the detection schema between features.
Preliminary results have shown that this mechanism
is effective in detecting features and achieves
a repeatability performance in line with current
state-of-the-art detection methods.",
  url          = "http://www.cat.uab.cat/Public/Publications/2009/MOV2009"
}