File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: CN: Channel Normalization for Point Cloud Recognition

TitleCN: Channel Normalization for Point Cloud Recognition
Authors
Keywords3D recognition
Point cloud
Object detection
Classification
Issue Date2020
PublisherSpringer. The Proceedings' web site is located at https://link.springer.com/conference/eccv
Citation
Proceedings of the 16th European Conference on Computer Vision (ECCV), Online, Glasgow, UK, 23-28 August 2020, pt X, p. 600-616 How to Cite?
AbstractIn 3D recognition, to fuse multi-scale structure information, existing methods apply hierarchical frameworks stacked by multiple fusion layers for integrating current relative locations with structure information from the previous level. In this paper, we deeply analyze these point recognition frameworks and present a factor, called difference ratio, to measure the influence of structure information among different levels on the final representation. We discover that structure information in deeper layers is overwhelmed by information in shallower layers in generating the final features, which prevents the model from understanding the point cloud in a global view. Inspired by this observation, we propose a novel channel normalization scheme to balance structure information among different layers and avoid excessive accumulation of shallow information, which benefits the model in exploiting and integrating multilayer structure information. We evaluate our channel normalization in several core 3D recognition tasks including classification, segmentation and detection. Experimental results show that our channel normalization further boosts the performance of state-of-the-art methods effectively.
Persistent Identifierhttp://hdl.handle.net/10722/294235
ISBN
Series/Report no.Lecture Notes in Computer Science ; vol. 12355

 

DC FieldValueLanguage
dc.contributor.authorYang, Z-
dc.contributor.authorSun, Y-
dc.contributor.authorLiu, S-
dc.contributor.authorQi, X-
dc.contributor.authorJia, J-
dc.date.accessioned2020-11-23T08:28:22Z-
dc.date.available2020-11-23T08:28:22Z-
dc.date.issued2020-
dc.identifier.citationProceedings of the 16th European Conference on Computer Vision (ECCV), Online, Glasgow, UK, 23-28 August 2020, pt X, p. 600-616-
dc.identifier.isbn9783030586065-
dc.identifier.urihttp://hdl.handle.net/10722/294235-
dc.description.abstractIn 3D recognition, to fuse multi-scale structure information, existing methods apply hierarchical frameworks stacked by multiple fusion layers for integrating current relative locations with structure information from the previous level. In this paper, we deeply analyze these point recognition frameworks and present a factor, called difference ratio, to measure the influence of structure information among different levels on the final representation. We discover that structure information in deeper layers is overwhelmed by information in shallower layers in generating the final features, which prevents the model from understanding the point cloud in a global view. Inspired by this observation, we propose a novel channel normalization scheme to balance structure information among different layers and avoid excessive accumulation of shallow information, which benefits the model in exploiting and integrating multilayer structure information. We evaluate our channel normalization in several core 3D recognition tasks including classification, segmentation and detection. Experimental results show that our channel normalization further boosts the performance of state-of-the-art methods effectively.-
dc.languageeng-
dc.publisherSpringer. The Proceedings' web site is located at https://link.springer.com/conference/eccv-
dc.relation.ispartofEuropean Conference on Computer Vision (ECCV)-
dc.relation.ispartofseriesLecture Notes in Computer Science ; vol. 12355-
dc.subject3D recognition-
dc.subjectPoint cloud-
dc.subjectObject detection-
dc.subjectClassification-
dc.titleCN: Channel Normalization for Point Cloud Recognition-
dc.typeConference_Paper-
dc.identifier.emailQi, X: xjqi@eee.hku.hk-
dc.identifier.authorityQi, X=rp02666-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-030-58607-2_35-
dc.identifier.hkuros320010-
dc.identifier.volumept X-
dc.identifier.spage600-
dc.identifier.epage616-
dc.publisher.placeCham-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats