File Download
Supplementary

Conference Paper: Context-Aware Spatio-Recurrent Curvilinear Structure Segmentation

TitleContext-Aware Spatio-Recurrent Curvilinear Structure Segmentation
Authors
Issue Date2019
PublisherIEEE / Computer Vision Foundation (CVF).
Citation
Proceedings of IEEE / CVF Conference on Computer Vision and Pattern Recognition 2019 (CVPR 2019). Long Beach, CA, USA, 16-20 June 2019, p. 12648-12657 How to Cite?
AbstractCurvilinear structures are frequently observed in various images in different forms, such as blood vessels or neuronal boundaries in biomedical images. In this paper, we propose a novel curvilinear structure segmentation approach using context-aware spatio-recurrent networks. Instead of directly segmenting the whole image or densely segmenting fixed-sized local patches, our method recurrently samples patches with varied scales from the target image with learned policy and processes them locally, which is similar to the behavior of changing retinal fixations in the human visual system and it is beneficial for capturing the multi- scale or hierarchical modality of the complex curvilinear structures. In specific, the policy of choosing local patches is attentively learned based on the contextual information of the image and the historical sampling experience. In this way, with more patches sampled and refined, the segmentation of the whole image can be progressively improved. To validate our approach, comparison experiments on different types of image data are conducted and the sampling procedures for exemplar images are illustrated. We demonstrate that our method achieves the state-of-the-art performance in public datasets.
DescriptionPaper no. 5904
Persistent Identifierhttp://hdl.handle.net/10722/273022

 

DC FieldValueLanguage
dc.contributor.authorWang, FGG-
dc.contributor.authorGu, Y-
dc.contributor.authorLiu, WX-
dc.contributor.authorYu, YL-
dc.contributor.authorHe, SF-
dc.contributor.authorPan, J-
dc.date.accessioned2019-08-06T09:21:04Z-
dc.date.available2019-08-06T09:21:04Z-
dc.date.issued2019-
dc.identifier.citationProceedings of IEEE / CVF Conference on Computer Vision and Pattern Recognition 2019 (CVPR 2019). Long Beach, CA, USA, 16-20 June 2019, p. 12648-12657-
dc.identifier.urihttp://hdl.handle.net/10722/273022-
dc.descriptionPaper no. 5904-
dc.description.abstractCurvilinear structures are frequently observed in various images in different forms, such as blood vessels or neuronal boundaries in biomedical images. In this paper, we propose a novel curvilinear structure segmentation approach using context-aware spatio-recurrent networks. Instead of directly segmenting the whole image or densely segmenting fixed-sized local patches, our method recurrently samples patches with varied scales from the target image with learned policy and processes them locally, which is similar to the behavior of changing retinal fixations in the human visual system and it is beneficial for capturing the multi- scale or hierarchical modality of the complex curvilinear structures. In specific, the policy of choosing local patches is attentively learned based on the contextual information of the image and the historical sampling experience. In this way, with more patches sampled and refined, the segmentation of the whole image can be progressively improved. To validate our approach, comparison experiments on different types of image data are conducted and the sampling procedures for exemplar images are illustrated. We demonstrate that our method achieves the state-of-the-art performance in public datasets.-
dc.languageeng-
dc.publisherIEEE / Computer Vision Foundation (CVF).-
dc.relation.ispartofThe IEEE Conference on Computer Vision and Pattern Recognition-
dc.titleContext-Aware Spatio-Recurrent Curvilinear Structure Segmentation-
dc.typeConference_Paper-
dc.identifier.emailPan, J: jpan@cs.hku.hk-
dc.identifier.authorityPan, J=rp01984-
dc.description.naturepostprint-
dc.identifier.hkuros300345-
dc.identifier.spage12648-
dc.identifier.epage12657-
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats