DC Field | Value | Language |
---|---|---|
dc.contributor.author | Shin, Ukcheol | ko |
dc.contributor.author | Park, Kwanyong | ko |
dc.contributor.author | Lee, Kyunghyun | ko |
dc.contributor.author | Lee, Byeong-Uk | ko |
dc.contributor.author | Kweon, In So | ko |
dc.date.accessioned | 2023-06-21T06:01:17Z | - |
dc.date.available | 2023-06-21T06:01:17Z | - |
dc.date.created | 2023-06-21 | - |
dc.date.issued | 2023-07 | - |
dc.identifier.citation | MACHINE VISION AND APPLICATIONS, v.34, no.4 | - |
dc.identifier.issn | 0932-8092 | - |
dc.identifier.uri | http://hdl.handle.net/10203/307403 | - |
dc.description.abstract | Depth estimation from thermal images is one potential solution to achieve reliability and robustness against diverse weather, lighting, and environmental conditions. Also, a self-supervised training method further boosts its scalability to various scenar-ios, which are usually impossible to collect ground-truth labels, such as GPS-denied and LiDAR-denied conditions. However, self-supervision from thermal images is usually insufficient to train networks due to the thermal image properties, such as low-contrast and textureless properties. Introducing additional self-supervision sources (e.g., RGB images) also introduces further hardware and software constraints, such as complicated multi-sensor calibration and synchronized data acquisition. Therefore, this manuscript proposes a novel training framework combining self-supervised learning and adversarial feature adaptation to leverage additional modality information without such constraints. The framework aims to train a network that estimates a monocular depth map from a thermal image in a self-supervised manner. In the training stage, the framework uti-lizes two self-supervisions; image reconstruction of unpaired RGB-thermal images and adversarial feature adaptation between unpaired RGB-thermal features. Based on the proposed method, the trained network achieves state-of-the-art quantitative results and edge-preserved depth estimation results compared to previous methods. Our source code is available at www. github.com/ukcheolshin/SelfDepth4Thermal | - |
dc.language | English | - |
dc.publisher | SPRINGER | - |
dc.title | Joint self-supervised learning and adversarial adaptation for monocular depth estimation from thermal image | - |
dc.type | Article | - |
dc.identifier.wosid | 001000235500001 | - |
dc.identifier.scopusid | 2-s2.0-85160934788 | - |
dc.type.rims | ART | - |
dc.citation.volume | 34 | - |
dc.citation.issue | 4 | - |
dc.citation.publicationname | MACHINE VISION AND APPLICATIONS | - |
dc.identifier.doi | 10.1007/s00138-023-01404-3 | - |
dc.contributor.localauthor | Kweon, In So | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Depth estimation | - |
dc.subject.keywordAuthor | Self-supervised learning | - |
dc.subject.keywordAuthor | Adversarial domain adaptation | - |
dc.subject.keywordAuthor | Thermal image | - |
dc.subject.keywordAuthor | Thermal vision | - |
dc.subject.keywordPlus | VISION | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.