Generalized Adaptive Dictionary Learning via Domain Shift Minimization
This addresses domain adaptation problems in visual recognition, offering an incremental improvement for tasks with contrasting data sources.
The paper tackles the challenge of object recognition when training and test data come from different domains by proposing a method to learn a common adaptive dictionary that minimizes domain shift and preserves data structure. Experiments on image classification show it outperforms existing state-of-the-art methods.
Visual data driven dictionaries have been successfully employed for various object recognition and classification tasks. However, the task becomes more challenging if the training and test data are from contrasting domains. In this paper, we propose a novel and generalized approach towards learning an adaptive and common dictionary for multiple domains. Precisely, we project the data from different domains onto a low dimensional space while preserving the intrinsic structure of data from each domain. We also minimize the domain-shift among the data from each pair of domains. Simultaneously, we learn a common adaptive dictionary. Our algorithm can also be modified to learn class-specific dictionaries which can be used for classification. We additionally propose a discriminative manifold regularization which imposes the intrinsic structure of class specific features onto the sparse coefficients. Experiments on image classification show that our approach fares better compared to the existing state-of-the-art methods.