Multi-resolution distillation for self-supervised monocular depth estimation

Cited 1 time in webofscience Cited 0 time in scopus
  • Hit : 86
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorLee, Sebinko
dc.contributor.authorIm, Woobinko
dc.contributor.authorYoon, Sung-Euiko
dc.date.accessioned2023-12-09T05:00:21Z-
dc.date.available2023-12-09T05:00:21Z-
dc.date.created2023-12-08-
dc.date.created2023-12-08-
dc.date.created2023-12-08-
dc.date.created2023-12-08-
dc.date.issued2023-12-
dc.identifier.citationPATTERN RECOGNITION LETTERS, v.176, pp.215 - 222-
dc.identifier.issn0167-8655-
dc.identifier.urihttp://hdl.handle.net/10203/316116-
dc.description.abstractObtaining dense depth ground-truth is not trivial, which leads to the introduction of self-supervised monocular depth estimation. Most self-supervised methods utilize the photometric loss as the primary supervisory signal to optimize a depth network. However, such self-supervised training often falls into an undesirable local minimum due to the ambiguity of the photometric loss. In this paper, we propose a novel self-distillation training scheme that provides a new self-supervision signal, depth consistency among different input resolutions, to the depth network. We further introduce a gradient masking strategy that adjusts the self-supervision signal of the depth consistency during back-propagation to boost the effectiveness of our depth consistency. Experiments demonstrate that our method brings meaningful performance improvements when applied to various depth network architectures. Furthermore, our method outperforms the existing self-supervised methods on KITTI, Cityscapes, and DrivingStereo datasets by a noteworthy margin. © 2023 Elsevier B.V.-
dc.languageEnglish-
dc.publisherELSEVIER-
dc.titleMulti-resolution distillation for self-supervised monocular depth estimation-
dc.typeArticle-
dc.identifier.wosid001160579300001-
dc.identifier.scopusid2-s2.0-85177224431-
dc.type.rimsART-
dc.citation.volume176-
dc.citation.beginningpage215-
dc.citation.endingpage222-
dc.citation.publicationnamePATTERN RECOGNITION LETTERS-
dc.identifier.doi10.1016/j.patrec.2023.11.001-
dc.contributor.localauthorYoon, Sung-Eui-
dc.description.isOpenAccessN-
dc.type.journalArticleArticle-
dc.subject.keywordAuthorDeep learning-
dc.subject.keywordAuthorMonocular depth estimation-
dc.subject.keywordAuthorSelf-distillation-
dc.subject.keywordAuthorSelf-supervised learning-
Appears in Collection
CS-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 1 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0