Unsupervised Domain Adaptation Based on Subspace Alignment

• Published in: Domain Adaptation in Computer Vision Applications pp. 81 - 94
• Main Authors: Fernando, Basura, Aljundi, Rahaf, Emonet, Rémi, Habrard, Amaury, Sebban, Marc, Tuytelaars, Tinne
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2017; Springer International Publishing
• Series: Advances in Computer Vision and Pattern Recognition
• Subjects: Artificial intelligence; Image processing; Public administration
• ISBN: 3319583468; 9783319583464
• ISSN: 2191-6586; 2191-6594
• DOI: 10.1007/978-3-319-58347-1_4

Subspace-based domain adaptation methods have been very successful in the context of image recognition. In this chapter, we discuss methods using Subspace Alignment (SA). They are based on a mapping function which aligns the source subspace with the target one, so as to obtain a domain invariant feature space. The solution of the corresponding optimization problem can be obtained in closed form, leading to a simple to implement and fast algorithm. The only hyperparameter involved corresponds to the dimension of the subspaces. We give two methods, SA and SA-MLE, for setting this variable. SA is a purely linear method. As a nonlinear extension, Landmarks-based Kernelized Subspace alignment (SA)Landmarks-based kernelized SubSpace Alignment (LSSA) Alignment (LSSA) first projects the data nonlinearly based on a set of landmarks, which have been selected so as to reduce the Discrepancy between the domains.