Domain adaptation using weighted sub-space sampling for object categorization

This paper describes a method of cross-domain object categorization, using the concept of domain adaptation. Here, a classifier is trained using samples from the source/auxiliary domain and performance is observed on a set of test samples taken from a different domain, termed as the target domain. To overcome the difference between the two domains, we aim to find a sequence of optimally weighted sub-spaces, lying on the geodesic path on Grassmann manifold, such that the instances from both the domains follow similar distributions when projected onto the sub-spaces. Hence, the method models the gradual change of the distribution of data from source to target domain, using a sequence of weighted sub-spaces. Results show that the proposed method of unsupervised domain adaptation provides better classification accuracy than a few state of the art methods.