FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308017823> ?p ?o ?g. }

• W4308017823 endingPage "2343" @default.
• W4308017823 startingPage "2343" @default.
• W4308017823 abstract "Although many recent deep learning methods have achieved good performance in point cloud analysis, most of them are built upon the heavy cost of manual labeling. Unsupervised representation learning methods have attracted increasing attention due to their high label efficiency. How to learn more useful representations from unlabeled 3D point clouds is still a challenging problem. Addressing this problem, we propose a novel unsupervised learning approach for point cloud analysis, named ULD-Net, consisting of an equivariant-crop (equiv-crop) module to achieve dense similarity learning. We propose dense similarity learning that maximizes consistency across two randomly transformed global–local views at both the instance level and the point level. To build feature correspondence between global and local views, an equiv-crop is proposed to transform features from the global scope to the local scope. Unlike previous methods that require complicated designs, such as negative pairs and momentum encoders, our ULD-Net benefits from the simple Siamese network that relies solely on stop-gradient operation preventing the network from collapsing. We also utilize the feature separability constraint for more representative embeddings. Experimental results show that our ULD-Net achieves the best results of context-based unsupervised methods and comparable performances to supervised models in shape classification and segmentation tasks. On the linear support vector machine classification benchmark, our ULD-Net surpasses the best context-based method spatiotemporal self-supervised representation learning (STRL) by 1.1% overall accuracy. On tasks with fine-tuning, our ULD-Net outperforms STRL under fully supervised and semisupervised settings, in particular, 0.1% accuracy gain on the ModelNet40 classification benchmark, and 0.6% medium intersection of union gain on the ShapeNet part segmentation benchmark." @default.
• W4308017823 created "2022-11-07" @default.
• W4308017823 creator A5009095415 @default.
• W4308017823 creator A5021263022 @default.
• W4308017823 creator A5042607903 @default.
• W4308017823 creator A5042731554 @default.
• W4308017823 creator A5048813928 @default.
• W4308017823 creator A5050842662 @default.
• W4308017823 creator A5070377754 @default.
• W4308017823 creator A5076512588 @default.
• W4308017823 date "2022-11-29" @default.
• W4308017823 modified "2023-10-15" @default.
• W4308017823 title "ULD-Net: 3D unsupervised learning by dense similarity learning with equivariant-crop" @default.
• W4308017823 cites W2038349953 @default.
• W4308017823 cites W2553307952 @default.
• W4308017823 cites W2564703108 @default.
• W4308017823 cites W2963312728 @default.
• W4308017823 cites W2979750740 @default.
• W4308017823 cites W3009750677 @default.
• W4308017823 cites W3153465022 @default.
• W4308017823 cites W3178738710 @default.
• W4308017823 cites W4205749550 @default.
• W4308017823 doi "https://doi.org/10.1364/josaa.473657" @default.
• W4308017823 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36520758" @default.
• W4308017823 hasPublicationYear "2022" @default.
• W4308017823 type Work @default.
• W4308017823 citedByCount "0" @default.
• W4308017823 crossrefType "journal-article" @default.
• W4308017823 hasAuthorship W4308017823A5009095415 @default.
• W4308017823 hasAuthorship W4308017823A5021263022 @default.
• W4308017823 hasAuthorship W4308017823A5042607903 @default.
• W4308017823 hasAuthorship W4308017823A5042731554 @default.
• W4308017823 hasAuthorship W4308017823A5048813928 @default.
• W4308017823 hasAuthorship W4308017823A5050842662 @default.
• W4308017823 hasAuthorship W4308017823A5070377754 @default.
• W4308017823 hasAuthorship W4308017823A5076512588 @default.
• W4308017823 hasConcept C103278499 @default.
• W4308017823 hasConcept C108583219 @default.
• W4308017823 hasConcept C115961682 @default.
• W4308017823 hasConcept C119857082 @default.
• W4308017823 hasConcept C131979681 @default.
• W4308017823 hasConcept C13280743 @default.
• W4308017823 hasConcept C138885662 @default.
• W4308017823 hasConcept C151730666 @default.
• W4308017823 hasConcept C153180895 @default.
• W4308017823 hasConcept C154945302 @default.
• W4308017823 hasConcept C17744445 @default.
• W4308017823 hasConcept C185798385 @default.
• W4308017823 hasConcept C199539241 @default.
• W4308017823 hasConcept C205649164 @default.
• W4308017823 hasConcept C2776359362 @default.
• W4308017823 hasConcept C2776401178 @default.
• W4308017823 hasConcept C2779343474 @default.
• W4308017823 hasConcept C41008148 @default.
• W4308017823 hasConcept C41895202 @default.
• W4308017823 hasConcept C59404180 @default.
• W4308017823 hasConcept C8038995 @default.
• W4308017823 hasConcept C86803240 @default.
• W4308017823 hasConcept C94625758 @default.
• W4308017823 hasConceptScore W4308017823C103278499 @default.
• W4308017823 hasConceptScore W4308017823C108583219 @default.
• W4308017823 hasConceptScore W4308017823C115961682 @default.
• W4308017823 hasConceptScore W4308017823C119857082 @default.
• W4308017823 hasConceptScore W4308017823C131979681 @default.
• W4308017823 hasConceptScore W4308017823C13280743 @default.
• W4308017823 hasConceptScore W4308017823C138885662 @default.
• W4308017823 hasConceptScore W4308017823C151730666 @default.
• W4308017823 hasConceptScore W4308017823C153180895 @default.
• W4308017823 hasConceptScore W4308017823C154945302 @default.
• W4308017823 hasConceptScore W4308017823C17744445 @default.
• W4308017823 hasConceptScore W4308017823C185798385 @default.
• W4308017823 hasConceptScore W4308017823C199539241 @default.
• W4308017823 hasConceptScore W4308017823C205649164 @default.
• W4308017823 hasConceptScore W4308017823C2776359362 @default.
• W4308017823 hasConceptScore W4308017823C2776401178 @default.
• W4308017823 hasConceptScore W4308017823C2779343474 @default.
• W4308017823 hasConceptScore W4308017823C41008148 @default.
• W4308017823 hasConceptScore W4308017823C41895202 @default.
• W4308017823 hasConceptScore W4308017823C59404180 @default.
• W4308017823 hasConceptScore W4308017823C8038995 @default.
• W4308017823 hasConceptScore W4308017823C86803240 @default.
• W4308017823 hasConceptScore W4308017823C94625758 @default.
• W4308017823 hasFunder F4320321543 @default.
• W4308017823 hasFunder F4320335777 @default.
• W4308017823 hasIssue "12" @default.
• W4308017823 hasLocation W43080178231 @default.
• W4308017823 hasLocation W43080178232 @default.
• W4308017823 hasOpenAccess W4308017823 @default.
• W4308017823 hasPrimaryLocation W43080178231 @default.
• W4308017823 hasRelatedWork W2546942002 @default.
• W4308017823 hasRelatedWork W2597787948 @default.
• W4308017823 hasRelatedWork W3123344745 @default.
• W4308017823 hasRelatedWork W3192794374 @default.
• W4308017823 hasRelatedWork W3208584567 @default.
• W4308017823 hasRelatedWork W4221031031 @default.
• W4308017823 hasRelatedWork W4221136938 @default.
• W4308017823 hasRelatedWork W4246751904 @default.
• W4308017823 hasRelatedWork W4302303815 @default.

• next