Improving object detection by removing noisy samples from training sets.
Gunawan HermanGetian YeJie XuBang ZhangPublished in: Multimedia Information Retrieval (2008)
Keyphrases
- object detection
- training set
- training samples
- data samples
- training examples
- data sets
- multi class
- active learning
- classification accuracy
- computer vision
- object classification
- noisy data
- feature space
- supervised learning
- background subtraction
- face images
- test set
- object categories
- pedestrian detection
- pairwise
- object recognition
- feature extraction
- svm classifier
- decision trees
- human detection
- feature selection