Unsupervised Open Domain Recognition by Semantic Discrepancy Minimization
This addresses a challenging problem in domain adaptation for computer vision, where unknown categories in the target domain complicate classification, but the approach is incremental as it builds on existing graph CNN and discrepancy minimization techniques.
The paper tackles unsupervised open domain recognition, where labeled source categories are a subset of unlabeled target categories, by proposing a method to classify both known and unknown categories in the target domain. It introduces Semantic-Guided Matching Discrepancy and a limited balance constraint, achieving superior results in recognizing images of both known and unknown categories.
We address the unsupervised open domain recognition (UODR) problem, where categories in labeled source domain S is only a subset of those in unlabeled target domain T. The task is to correctly classify all samples in T including known and unknown categories. UODR is challenging due to the domain discrepancy, which becomes even harder to bridge when a large number of unknown categories exist in T. Moreover, the classification rules propagated by graph CNN (GCN) may be distracted by unknown categories and lack generalization capability. To measure the domain discrepancy for asymmetric label space between S and T, we propose Semantic-Guided Matching Discrepancy (SGMD), which first employs instance matching between S and T, and then the discrepancy is measured by a weighted feature distance between matched instances. We further design a limited balance constraint to achieve a more balanced classification output on known and unknown categories. We develop Unsupervised Open Domain Transfer Network (UODTN), which learns both the backbone classification network and GCN jointly by reducing the SGMD, enforcing the limited balance constraint and minimizing the classification loss on S. UODTN better preserves the semantic structure and enforces the consistency between the learned domain invariant visual features and the semantic embeddings. Experimental results show superiority of our method on recognizing images of both known and unknown categories.