FIDOS: A generalized Fisher based feature extraction method for domain shift

Traditional pattern recognition techniques often assume that the data sets used for training and testing follow the same distribution. However, this assumption is usually not true for many real world problems as data from the same classes but different domains, e.g., data are collected under different conditions, may show different characteristics. We introduce FIDOS, a generalized FIsher based method for DOmain Shift problem, that aims at learning invariant features across domains in a supervised manner.Different from classical Fisher feature extraction, FIDOS aims to minimize not only the within-class scatter but also the difference in distributions between domains. Therefore, the subspace constructed by FIDOS reduces the drift in distributions among different domains and at the same time preserves the discriminants across classes. Another advantage of FIDOS over classical Fisher is that FIDOS extracts more features when multiple source domains are available in the training set; this is essential for a good classification especially when the number of classes is small. Experimental results on both artificial and real data and comparisons with other methods demonstrate the efficiency of our method in classifying objects under domain shift situations.

► We present FIDOS, a generalized Fisher feature extraction, to deal with domain shift.
► FIDOS reduces drift between domains while preserving discriminant among classes.
► FIDOS handles both single and multiple source domain scenarios.
► FIDOS, in general, extracts more features than classical Fisher does.
► Like Fisher, FIDOS is invariant to linear transformation and computationally efficient.