Using k-nearest neighbors to handle missing weak classifiers in a boosted cascade.
Pierre BougesThierry ChateauChristophe BlancGaëlle LoosliPublished in: ICPR (2012)
Keyphrases
- k nearest neighbor
- weak classifiers
- knn
- boosted classifiers
- nearest neighbor
- training set
- linear combination
- cascade classifier
- face detection
- face detector
- training samples
- classification method
- classification error
- image patches
- classification algorithm
- neural network
- support vector machine svm
- object detection
- object class
- mean shift
- text classification
- distance function
- support vector machine
- feature selection
- data sets
- missing values
- multiscale