Viewpoint-Invariant Learning and Detection of Human Heads
We present a method to learn models of human heads for
the purpose of detection from different viewing angles. We
focus on a model where objects are represented as constel-
lations of rigid features (parts). Variability is represented
by a joint probability density function (pdf) on the shape of
the constellation. In a first stage, the method automatically
identifies distinctive features in the training set using an in-
terest operator followed by vector quantization. The set of
model parameters, including the shape pdf, is then learned
using expectation maximization. Experiments show good
generalization performance to novel viewpoints and unseen
faces. Performance is above 90% correct with less than 1s
computation time per image.
Download: pdf
Text Reference
Markus Weber, Wolfgang Einhaeuser, Max Welling, and Pietro Perona. Viewpoint-invariant learning and detection of human heads. In AFGR, 20–27. 2000.BibTeX Reference
@inproceedings{WeberEWP_AFGR_2000,AUTHOR = "Weber, Markus and Einhaeuser, Wolfgang and Welling, Max and Perona, Pietro",
TAG = "object_recognition",
TITLE = "Viewpoint-Invariant Learning and Detection of Human Heads",
BOOKTITLE = "AFGR",
YEAR = "2000",
PAGES = "20-27",
BIBSOURCE = "http://www.visionbib.com/bibliography/people903.html#TT70425"
}