Bayes error

A Novel Approach to Expression Recognition from Non-frontal Face Images

We propose a novel approach to expression recognition from non-frontal view facial images. Under the unified Bayes theoretical framework, the recognition problem can be formulated as an optimization problem of minimizing an upper bound of Bayes error.

Optimizing Multi-class Spatio-Spectral Filters via Bayes Error Estimation for EEG Classification

We generalize the two-class common spatio-spectral patterns (CSSPs) method to multi-class cases on the electroencephalogram (EEG) classification.