Matches in SemOpenAlex for { <https://semopenalex.org/work/W2955873422> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2955873422 abstract "We propose a new supervized learning framework for oversegmenting 3D point clouds into superpoints. We cast this problem as learning deep embeddings of the local geometry and radiometry of 3D points, such that the border of objects presents high contrasts. The embeddings are computed using a lightweight neural network operating on the points' local neighborhood. Finally, we formulate point cloud oversegmentation as a graph partition problem with respect to the learned embeddings. This new approach allows us to set a new state-of-the-art in point cloud oversegmentation by a significant margin, on a dense indoor dataset (S3DIS) and a sparse outdoor one (vKITTI). Our best solution requires over five times fewer superpoints to reach similar performance than previously published methods on S3DIS. Furthermore, we show that our framework can be used to improve superpoint-based semantic segmentation algorithms, setting a new state-of-the-art for this task as well." @default.
- W2955873422 created "2019-07-12" @default.
- W2955873422 creator A5033114563 @default.
- W2955873422 creator A5040100965 @default.
- W2955873422 date "2019-06-01" @default.
- W2955873422 modified "2023-10-17" @default.
- W2955873422 title "Point Cloud Oversegmentation With Graph-Structured Deep Metric Learning" @default.
- W2955873422 cites W1950112384 @default.
- W2955873422 cites W1967535109 @default.
- W2955873422 cites W1975517671 @default.
- W2955873422 cites W1999478155 @default.
- W2955873422 cites W2030346542 @default.
- W2955873422 cites W2058849570 @default.
- W2955873422 cites W2065742895 @default.
- W2955873422 cites W2072723786 @default.
- W2955873422 cites W2105910985 @default.
- W2955873422 cites W2107023258 @default.
- W2955873422 cites W2118246710 @default.
- W2955873422 cites W2119531662 @default.
- W2955873422 cites W2135249503 @default.
- W2955873422 cites W2135431554 @default.
- W2955873422 cites W2157364932 @default.
- W2955873422 cites W2194775991 @default.
- W2955873422 cites W2221409856 @default.
- W2955873422 cites W2346698422 @default.
- W2955873422 cites W2556802233 @default.
- W2955873422 cites W2585293115 @default.
- W2955873422 cites W2606202972 @default.
- W2955873422 cites W2620650400 @default.
- W2955873422 cites W2804865177 @default.
- W2955873422 cites W2804872164 @default.
- W2955873422 cites W2883606943 @default.
- W2955873422 cites W2949699301 @default.
- W2955873422 cites W2963281829 @default.
- W2955873422 cites W2963775347 @default.
- W2955873422 cites W2963988212 @default.
- W2955873422 cites W2964206727 @default.
- W2955873422 cites W3103830808 @default.
- W2955873422 cites W3104148512 @default.
- W2955873422 cites W3122159272 @default.
- W2955873422 cites W4210880854 @default.
- W2955873422 doi "https://doi.org/10.1109/cvpr.2019.00762" @default.
- W2955873422 hasPublicationYear "2019" @default.
- W2955873422 type Work @default.
- W2955873422 sameAs 2955873422 @default.
- W2955873422 citedByCount "97" @default.
- W2955873422 countsByYear W29558734222019 @default.
- W2955873422 countsByYear W29558734222020 @default.
- W2955873422 countsByYear W29558734222021 @default.
- W2955873422 countsByYear W29558734222022 @default.
- W2955873422 countsByYear W29558734222023 @default.
- W2955873422 crossrefType "proceedings-article" @default.
- W2955873422 hasAuthorship W2955873422A5033114563 @default.
- W2955873422 hasAuthorship W2955873422A5040100965 @default.
- W2955873422 hasBestOaLocation W29558734222 @default.
- W2955873422 hasConcept C111919701 @default.
- W2955873422 hasConcept C127413603 @default.
- W2955873422 hasConcept C131979681 @default.
- W2955873422 hasConcept C132525143 @default.
- W2955873422 hasConcept C154945302 @default.
- W2955873422 hasConcept C176217482 @default.
- W2955873422 hasConcept C21547014 @default.
- W2955873422 hasConcept C41008148 @default.
- W2955873422 hasConcept C79974875 @default.
- W2955873422 hasConcept C80444323 @default.
- W2955873422 hasConceptScore W2955873422C111919701 @default.
- W2955873422 hasConceptScore W2955873422C127413603 @default.
- W2955873422 hasConceptScore W2955873422C131979681 @default.
- W2955873422 hasConceptScore W2955873422C132525143 @default.
- W2955873422 hasConceptScore W2955873422C154945302 @default.
- W2955873422 hasConceptScore W2955873422C176217482 @default.
- W2955873422 hasConceptScore W2955873422C21547014 @default.
- W2955873422 hasConceptScore W2955873422C41008148 @default.
- W2955873422 hasConceptScore W2955873422C79974875 @default.
- W2955873422 hasConceptScore W2955873422C80444323 @default.
- W2955873422 hasLocation W29558734221 @default.
- W2955873422 hasLocation W29558734222 @default.
- W2955873422 hasLocation W29558734223 @default.
- W2955873422 hasLocation W29558734224 @default.
- W2955873422 hasOpenAccess W2955873422 @default.
- W2955873422 hasPrimaryLocation W29558734221 @default.
- W2955873422 hasRelatedWork W2048509577 @default.
- W2955873422 hasRelatedWork W2275108558 @default.
- W2955873422 hasRelatedWork W2383532021 @default.
- W2955873422 hasRelatedWork W266446692 @default.
- W2955873422 hasRelatedWork W275032887 @default.
- W2955873422 hasRelatedWork W2919233342 @default.
- W2955873422 hasRelatedWork W2962771259 @default.
- W2955873422 hasRelatedWork W3107474891 @default.
- W2955873422 hasRelatedWork W4236152845 @default.
- W2955873422 hasRelatedWork W4312098327 @default.
- W2955873422 isParatext "false" @default.
- W2955873422 isRetracted "false" @default.
- W2955873422 magId "2955873422" @default.
- W2955873422 workType "article" @default.