DC Field | Value | Language |
---|---|---|
dc.contributor.author | 김남엽 | - |
dc.contributor.author | Son, Taeyoung | - |
dc.contributor.author | 박재현 | - |
dc.contributor.author | Lan, Cuiling | - |
dc.contributor.author | Zeng, Wenjun | - |
dc.contributor.author | 곽수하 | - |
dc.date.accessioned | 2024-03-07T00:24:35Z | - |
dc.date.available | 2024-03-07T00:24:35Z | - |
dc.date.created | 2024-03-06 | - |
dc.date.issued | 2023-06-01 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/122811 | - |
dc.description.abstract | Domain generalization for semantic segmentation is highly demanded in real applications, where a trained model is expected to work well in previously unseen domains. One challenge lies in the lack of data which could cover the diverse distributions of the possible unseen domains for training. In this paper, we propose a WEb-image assisted Domain GEneralization (WEDGE) scheme, which is the first to exploit the diversity of web-crawled images for generalizable semantic segmentation. To explore and exploit the real-world data distributions, we collect web-crawled images which present large diversity in terms of weather conditions, sites, lighting, camera styles, etc. We also present a method which injects styles of the web-crawled images into training images on-the-fly during training, which enables the network to experience images of diverse styles with reliable labels for effective training. Moreover, we use the web-crawled images with their predicted pseudo labels for training to further enhance the capability of the network. Extensive experiments demonstrate that our method clearly outperforms existing domain generalization techniques. | - |
dc.language | English | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.relation.isPartOf | 2023 IEEE International Conference on Robotics and Automation, ICRA 2023 | - |
dc.relation.isPartOf | Proceedings - IEEE International Conference on Robotics and Automation | - |
dc.title | WEDGE: Web-Image Assisted Domain Generalization for Semantic Segmentation | - |
dc.type | Conference | - |
dc.type.rims | CONF | - |
dc.identifier.bibliographicCitation | 2023 IEEE International Conference on Robotics and Automation, ICRA 2023, pp.9281 - 9288 | - |
dc.citation.conferenceDate | 2023-05-29 | - |
dc.citation.conferencePlace | UK | - |
dc.citation.endPage | 9288 | - |
dc.citation.startPage | 9281 | - |
dc.citation.title | 2023 IEEE International Conference on Robotics and Automation, ICRA 2023 | - |
dc.contributor.affiliatedAuthor | 김남엽 | - |
dc.contributor.affiliatedAuthor | 박재현 | - |
dc.contributor.affiliatedAuthor | 곽수하 | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.