File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Domain-Invariant Stereo Matching Networks

TitleDomain-Invariant Stereo Matching Networks
Authors
Issue Date2020
PublisherSpringer.
Citation
Proceedings of the 16th European Conference on Computer Vision (ECCV), Online, Glasgow, UK, 23-28 August 2020, pt 2, p. 420-439 How to Cite?
AbstractState-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel “domain normalization” approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) an end-to-end trainable structure-preserving graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep neural network models (e.g. MC-CNN [61]) fine-tuned with test-domain data.
Persistent Identifierhttp://hdl.handle.net/10722/294710
ISBN
Series/Report no.Lecture Notes in Computer Science (LNCS), v. 12347

 

DC FieldValueLanguage
dc.contributor.authorZhang, F-
dc.contributor.authorQi, X-
dc.contributor.authorYang, R-
dc.contributor.authorPrisacariu, V-
dc.contributor.authorWah, B-
dc.contributor.authorTorr, P-
dc.date.accessioned2020-12-08T07:40:45Z-
dc.date.available2020-12-08T07:40:45Z-
dc.date.issued2020-
dc.identifier.citationProceedings of the 16th European Conference on Computer Vision (ECCV), Online, Glasgow, UK, 23-28 August 2020, pt 2, p. 420-439-
dc.identifier.isbn9783030585358-
dc.identifier.urihttp://hdl.handle.net/10722/294710-
dc.description.abstractState-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel “domain normalization” approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) an end-to-end trainable structure-preserving graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep neural network models (e.g. MC-CNN [61]) fine-tuned with test-domain data.-
dc.languageeng-
dc.publisherSpringer.-
dc.relation.ispartofEuropean Conference on Computer Vision (ECCV) 2020-
dc.relation.ispartofseriesLecture Notes in Computer Science (LNCS), v. 12347-
dc.titleDomain-Invariant Stereo Matching Networks-
dc.typeConference_Paper-
dc.identifier.emailQi, X: xjqi@eee.hku.hk-
dc.identifier.authorityQi, X=rp02666-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-030-58536-5_25-
dc.identifier.hkuros320335-
dc.identifier.volumept 2-
dc.identifier.spage420-
dc.identifier.epage439-
dc.publisher.placeCham-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats