Ensemble of Multi-channel CNNs for Multi-class Time-Series Classification. Depth-Based Human Activity Recognition.
Jacek TrelinskiBogdan KwolekPublished in: ACIIDS (1) (2020)
Keyphrases
- multi class
- multi channel
- human activity recognition
- base classifiers
- activity recognition
- human activities
- binary classification problems
- accurate classifiers
- feature selection
- class probabilities
- single channel
- pairwise
- binary classification
- multi class classification
- support vector machine
- random forests
- multiple classes
- visual surveillance
- cost sensitive
- ensemble methods
- error correcting output codes
- multi class boosting
- neural network
- sampling methods
- boosting algorithms
- object detection
- multi class problems
- multi class classifier
- text mining
- training set
- object recognition
- decision trees
- image processing
- learning algorithm
- data sets