Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312788649> ?p ?o ?g. }
- W4312788649 endingPage "619" @default.
- W4312788649 startingPage "600" @default.
- W4312788649 abstract "Labelling point clouds fully is highly time-consuming and costly. As larger point cloud datasets with billions of points become more common, we ask whether the full annotation is even necessary, demonstrating that existing baselines designed under a fully annotated assumption only degrade slightly even when faced with 1% random point annotations. However, beyond this point, e.g., at 0.1% annotations, segmentation accuracy is unacceptably low. We observe that, as point clouds are samples of the 3D world, the distribution of points in a local neighbourhood is relatively homogeneous, exhibiting strong semantic similarity. Motivated by this, we propose a new weak supervision method to implicitly augment highly sparse supervision signals. Extensive experiments demonstrate the proposed Semantic Query Network (SQN) achieves promising performance on seven large-scale open datasets under weak supervision schemes, while requiring only 0.1% randomly annotated points for training, greatly reducing annotation cost and effort." @default.
- W4312788649 created "2023-01-05" @default.
- W4312788649 creator A5005662573 @default.
- W4312788649 creator A5010637110 @default.
- W4312788649 creator A5032533885 @default.
- W4312788649 creator A5036921959 @default.
- W4312788649 creator A5048108135 @default.
- W4312788649 creator A5060183988 @default.
- W4312788649 creator A5085971943 @default.
- W4312788649 date "2022-01-01" @default.
- W4312788649 modified "2023-10-18" @default.
- W4312788649 title "SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds" @default.
- W4312788649 cites W172584396 @default.
- W4312788649 cites W1903029394 @default.
- W4312788649 cites W2011674654 @default.
- W4312788649 cites W2135249503 @default.
- W4312788649 cites W2194775991 @default.
- W4312788649 cites W2295107390 @default.
- W4312788649 cites W2594519801 @default.
- W4312788649 cites W2775216572 @default.
- W4312788649 cites W2788158258 @default.
- W4312788649 cites W2797997528 @default.
- W4312788649 cites W2887121482 @default.
- W4312788649 cites W2888754481 @default.
- W4312788649 cites W2895472109 @default.
- W4312788649 cites W2960986959 @default.
- W4312788649 cites W2962912109 @default.
- W4312788649 cites W2963125977 @default.
- W4312788649 cites W2963182550 @default.
- W4312788649 cites W2963226018 @default.
- W4312788649 cites W2963231572 @default.
- W4312788649 cites W2963281829 @default.
- W4312788649 cites W2963517242 @default.
- W4312788649 cites W2963706542 @default.
- W4312788649 cites W2968557240 @default.
- W4312788649 cites W2979750740 @default.
- W4312788649 cites W2981199548 @default.
- W4312788649 cites W2981983525 @default.
- W4312788649 cites W2986037208 @default.
- W4312788649 cites W2990613095 @default.
- W4312788649 cites W2991216808 @default.
- W4312788649 cites W2991284067 @default.
- W4312788649 cites W2997890711 @default.
- W4312788649 cites W2997951504 @default.
- W4312788649 cites W3003437478 @default.
- W4312788649 cites W3010797203 @default.
- W4312788649 cites W3011788244 @default.
- W4312788649 cites W3012494314 @default.
- W4312788649 cites W3014902535 @default.
- W4312788649 cites W3025468529 @default.
- W4312788649 cites W3034239841 @default.
- W4312788649 cites W3034591723 @default.
- W4312788649 cites W3035275207 @default.
- W4312788649 cites W3035426492 @default.
- W4312788649 cites W3035524453 @default.
- W4312788649 cites W3035739565 @default.
- W4312788649 cites W3036406207 @default.
- W4312788649 cites W3036560856 @default.
- W4312788649 cites W3039448353 @default.
- W4312788649 cites W3093434340 @default.
- W4312788649 cites W3107479685 @default.
- W4312788649 cites W3109154950 @default.
- W4312788649 cites W3109944402 @default.
- W4312788649 cites W3110503160 @default.
- W4312788649 cites W3116959466 @default.
- W4312788649 cites W3119708198 @default.
- W4312788649 cites W3120926185 @default.
- W4312788649 cites W3153635465 @default.
- W4312788649 cites W3166573884 @default.
- W4312788649 cites W3167736108 @default.
- W4312788649 cites W3169120894 @default.
- W4312788649 cites W3174692508 @default.
- W4312788649 cites W3176802407 @default.
- W4312788649 cites W3177330511 @default.
- W4312788649 cites W3177450673 @default.
- W4312788649 cites W3181190968 @default.
- W4312788649 cites W3202162859 @default.
- W4312788649 cites W3202349074 @default.
- W4312788649 cites W3202541643 @default.
- W4312788649 cites W3203089644 @default.
- W4312788649 cites W3203495808 @default.
- W4312788649 cites W3203978358 @default.
- W4312788649 cites W3206972718 @default.
- W4312788649 cites W4214755140 @default.
- W4312788649 doi "https://doi.org/10.1007/978-3-031-19812-0_35" @default.
- W4312788649 hasPublicationYear "2022" @default.
- W4312788649 type Work @default.
- W4312788649 citedByCount "16" @default.
- W4312788649 countsByYear W43127886492022 @default.
- W4312788649 countsByYear W43127886492023 @default.
- W4312788649 crossrefType "book-chapter" @default.
- W4312788649 hasAuthorship W4312788649A5005662573 @default.
- W4312788649 hasAuthorship W4312788649A5010637110 @default.
- W4312788649 hasAuthorship W4312788649A5032533885 @default.
- W4312788649 hasAuthorship W4312788649A5036921959 @default.
- W4312788649 hasAuthorship W4312788649A5048108135 @default.
- W4312788649 hasAuthorship W4312788649A5060183988 @default.
- W4312788649 hasAuthorship W4312788649A5085971943 @default.