Matches in SemOpenAlex for { <https://semopenalex.org/work/W4224256510> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4224256510 endingPage "108722" @default.
- W4224256510 startingPage "108722" @default.
- W4224256510 abstract "Despite recent improvements in analyzing large-scale 3D point clouds, several problems still exist: (a) segmentation models suffer from intra-class inconsistency and inter-class indistinction; (b) the existing methods ignore the inherent long-tailed class distribution of real-world 3D data. These problems result in unsatisfactory semantic segmentation predictions, especially in object adjacent areas. To handle these problems, this paper proposes a novel Adjacent areas Refinement Network (ARNet). Specifically, an Adjacent areas Refinement (AR) module is designed, which consists of two parallel attention blocks. Besides, our proposed attention blocks can process a large number of points (N∼105) with a slight increase in the computational complexity and time cost. Additionally, to deal with the inherent long-tailed class distribution in real-world 3D data, imbalance adjustment loss and occupancy regression loss are introduced. Based on this, the proposed network can handle the classification of both majority and minority classes, which is essential in distinguishing the ambiguous parts in large-scale 3D scenes. The proposed AR module and the loss functions can be easily integrated into the cutting-edge backbone networks, contributing to better performance in modeling semantic inter-dependencies and significantly improving the accuracy of the state-of-the-art semantic segmentation methods on indoor and outdoor scenes." @default.
- W4224256510 created "2022-04-26" @default.
- W4224256510 creator A5041712189 @default.
- W4224256510 creator A5082076121 @default.
- W4224256510 creator A5084218062 @default.
- W4224256510 date "2022-09-01" @default.
- W4224256510 modified "2023-09-25" @default.
- W4224256510 title "Paying attention for adjacent areas: Learning discriminative features for large-scale 3D scene segmentation" @default.
- W4224256510 cites W2901259189 @default.
- W4224256510 cites W2963706542 @default.
- W4224256510 cites W2979750740 @default.
- W4224256510 cites W3014902535 @default.
- W4224256510 cites W3039448353 @default.
- W4224256510 cites W3125817276 @default.
- W4224256510 cites W3153465022 @default.
- W4224256510 cites W3210492780 @default.
- W4224256510 doi "https://doi.org/10.1016/j.patcog.2022.108722" @default.
- W4224256510 hasPublicationYear "2022" @default.
- W4224256510 type Work @default.
- W4224256510 citedByCount "3" @default.
- W4224256510 countsByYear W42242565102023 @default.
- W4224256510 crossrefType "journal-article" @default.
- W4224256510 hasAuthorship W4224256510A5041712189 @default.
- W4224256510 hasAuthorship W4224256510A5082076121 @default.
- W4224256510 hasAuthorship W4224256510A5084218062 @default.
- W4224256510 hasConcept C111919701 @default.
- W4224256510 hasConcept C119857082 @default.
- W4224256510 hasConcept C121332964 @default.
- W4224256510 hasConcept C124101348 @default.
- W4224256510 hasConcept C131979681 @default.
- W4224256510 hasConcept C153180895 @default.
- W4224256510 hasConcept C154945302 @default.
- W4224256510 hasConcept C162307627 @default.
- W4224256510 hasConcept C2524010 @default.
- W4224256510 hasConcept C2777212361 @default.
- W4224256510 hasConcept C2778755073 @default.
- W4224256510 hasConcept C2781238097 @default.
- W4224256510 hasConcept C28719098 @default.
- W4224256510 hasConcept C33923547 @default.
- W4224256510 hasConcept C41008148 @default.
- W4224256510 hasConcept C62520636 @default.
- W4224256510 hasConcept C89600930 @default.
- W4224256510 hasConcept C97931131 @default.
- W4224256510 hasConcept C98045186 @default.
- W4224256510 hasConceptScore W4224256510C111919701 @default.
- W4224256510 hasConceptScore W4224256510C119857082 @default.
- W4224256510 hasConceptScore W4224256510C121332964 @default.
- W4224256510 hasConceptScore W4224256510C124101348 @default.
- W4224256510 hasConceptScore W4224256510C131979681 @default.
- W4224256510 hasConceptScore W4224256510C153180895 @default.
- W4224256510 hasConceptScore W4224256510C154945302 @default.
- W4224256510 hasConceptScore W4224256510C162307627 @default.
- W4224256510 hasConceptScore W4224256510C2524010 @default.
- W4224256510 hasConceptScore W4224256510C2777212361 @default.
- W4224256510 hasConceptScore W4224256510C2778755073 @default.
- W4224256510 hasConceptScore W4224256510C2781238097 @default.
- W4224256510 hasConceptScore W4224256510C28719098 @default.
- W4224256510 hasConceptScore W4224256510C33923547 @default.
- W4224256510 hasConceptScore W4224256510C41008148 @default.
- W4224256510 hasConceptScore W4224256510C62520636 @default.
- W4224256510 hasConceptScore W4224256510C89600930 @default.
- W4224256510 hasConceptScore W4224256510C97931131 @default.
- W4224256510 hasConceptScore W4224256510C98045186 @default.
- W4224256510 hasFunder F4320321001 @default.
- W4224256510 hasLocation W42242565101 @default.
- W4224256510 hasOpenAccess W4224256510 @default.
- W4224256510 hasPrimaryLocation W42242565101 @default.
- W4224256510 hasRelatedWork W1652783584 @default.
- W4224256510 hasRelatedWork W1990254706 @default.
- W4224256510 hasRelatedWork W2024160000 @default.
- W4224256510 hasRelatedWork W2061273563 @default.
- W4224256510 hasRelatedWork W2285052147 @default.
- W4224256510 hasRelatedWork W2510758617 @default.
- W4224256510 hasRelatedWork W2729514902 @default.
- W4224256510 hasRelatedWork W2743258233 @default.
- W4224256510 hasRelatedWork W2773500201 @default.
- W4224256510 hasRelatedWork W4287995534 @default.
- W4224256510 hasVolume "129" @default.
- W4224256510 isParatext "false" @default.
- W4224256510 isRetracted "false" @default.
- W4224256510 workType "article" @default.