%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2023/09.22.20.32
%2 sid.inpe.br/sibgrapi/2023/09.22.20.32.42
%T Open Set Domain Adaptation Methods in Deep Networks for Image Recognition
%D 2023
%A Alvarenga e Silva, Lucas Fernando,
%A Almeida, Jurandy,
%@affiliation Universidade Estadual de Campinas – UNICAMP
%@affiliation Universidade Federal de Săo Carlos – UFScar
%E Clua, Esteban Walter Gonzalez,
%E Körting, Thales Sehn,
%E Paulovich, Fernando Vieira,
%E Feris, Rogerio,
%B Conference on Graphics, Patterns and Images, 36 (SIBGRAPI)
%C Rio Grande, RS
%8 Nov. 06-09, 2023
%S Proceedings
%K open set domain adaptation, unsupervised domain adaptation, domain adaptation, deep learning.
%X Deep learning (DL) has revolutionized various fields through its remarkable capacity to learn from raw data. However, in uncontrolled environments like in the wild, the performance of these systems might degrade to some extent, especially with unlabeled datasets. Naive approaches train DL models on labeled datasets (source domains) that resemble the unlabeled test dataset (target domain), but nonetheless, this approach may not yield optimal results due to domain and category-shift problems. These issues have been the primary focus of Unsupervised Domain Adaptation (UDA) and Open Set Recognition research areas. To address the domain-shift problem, we introduced the Multi-Source Domain Alignment Layers (MS-DIAL), a structural solution for multi-source UDA. MS-DIAL aligns the source domains and the target domain at various levels of the feature space, individually achieving competitive results comparable to the state-of-the-art, and when combined with other UDA methods, it further enhances transferability by up to 30.64% in relative performance gains. Subsequently, we tackled the demanding setup of Open Set Domain Adaptation (OSDA), where both domain and category-shift issues coexist. Our proposed approach involves dealing with negatives, extracting a high-confidence set of unknown instances, and using them as a hard constraint to refine the classification boundaries of OSDA methods. We assessed our proposal in an extensive set of experiments, which achieved up to 5.8% of absolute performance gains.
%@language en
%3 silva13.pdf