Unknown-Oriented Learning for Open Set Domain Adaptation

Research output: Chapters, Conference Papers, Creative and Literary Works (RGC: 12, 32, 41, 45)32_Refereed conference paper (with ISBN/ISSN)peer-review

View graph of relations

Related Research Unit(s)

Detail(s)

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2022, 17th European Conference
Subtitle of host publicationProceedings
EditorsGabriel Brostow, Moustapha Cissé, Tal Hassner, Shai Avidan, Giovanni Maria Farinella
PublisherSpringer, Cham
Number of pages17
VolumePart XXXIII
ISBN (Print)9783031198267
Publication statusOnline published - 12 Dec 2022

Publication series

NameLecture Notes in Computer Science
Volume13693
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Title17th European Conference on Computer Vision (ECCV 2022)
PlaceIsrael
CityTel-Aviv
Period23 - 27 October 2022

Abstract

Open set domain adaptation (OSDA) aims to tackle the distribution shift of partially shared categories between the source and target domains, meanwhile identifying target samples non-appeared in source domain. The key issue behind this problem is to classify these various unseen samples as unknown category with the absent of relevant knowledge from the source domain. Though impressing performance, existing works neglect the complex semantic information and huge intra-category variation of unknown category, incapable of representing the complicated distribution. To overcome this, we propose a novel Unknown-Oriented Learning (UOL) framework for OSDA, and it is composed of three stages: true unknown excavation, false unknown suppression and known alignment. Specifically, to excavate the diverse semantic information in unknown category, the multi-unknown detector (MUD) equipped with weight discrepancy constraint is proposed in true unknown excavation. During false unknown suppression, Source-to-Target grAdient Graph (S2TAG) is constructed to select reliable target samples with the proposed super confidence criteria. Then, Target-to-Target grAdient Graph (T2TAG) exploits the geometric structure in gradient manifold to obtain confident pseudo labels for target data. At the last stage, known alignment, the known samples in the target domain are aligned with the source domain to alleviate the domain gap. Extensive experiments demonstrate the superiority of our method compared with state-of-the-art methods on three benchmarks.

Research Area(s)

  • Domain Adaptation, Open Set, Graph

Citation Format(s)

Unknown-Oriented Learning for Open Set Domain Adaptation. / Liu, Jie; Guo, Xiaoqing; Yuan, Yixuan.

Computer Vision – ECCV 2022, 17th European Conference: Proceedings. ed. / Gabriel Brostow; Moustapha Cissé; Tal Hassner; Shai Avidan; Giovanni Maria Farinella. Vol. Part XXXIII Springer, Cham, 2022. (Lecture Notes in Computer Science; Vol. 13693).

Research output: Chapters, Conference Papers, Creative and Literary Works (RGC: 12, 32, 41, 45)32_Refereed conference paper (with ISBN/ISSN)peer-review