FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3093220398> ?p ?o ?g. }

• W3093220398 abstract "Despite the previous success of object analysis, detecting and segmenting a large number of object categories with a long-tailed data distribution remains a challenging problem and is less investigated. For a large-vocabulary classifier, the chance of obtaining noisy logits is much higher, which can easily lead to a wrong recognition. In this paper, we exploit prior knowledge of the relations among object categories to cluster fine-grained classes into coarser parent classes, and construct a classification tree that is responsible for parsing an object instance into a fine-grained category via its parent class. In the classification tree, as the number of parent class nodes are significantly less, their logits are less noisy and can be utilized to suppress the wrong/noisy logits existed in the fine-grained class nodes. As the way to construct the parent class is not unique, we further build multiple trees to form a classification forest where each tree contributes its vote to the fine-grained classification. To alleviate the imbalanced learning caused by the long-tail phenomena, we propose a simple yet effective resampling method, NMS Resampling, to re-balance the data distribution. Our method, termed as Forest R-CNN, can serve as a plug-and-play module being applied to most object recognition models for recognizing more than 1000 categories. Extensive experiments are performed on the large vocabulary dataset LVIS. Compared with the Mask R-CNN baseline, the Forest R-CNN significantly boosts the performance with 11.5% and 3.9% AP improvements on the rare categories and overall categories, respectively. Moreover, we achieve state-of-the-art results on the LVIS dataset. Code is available at https://github.com/JialianW/Forest_RCNN." @default.
• W3093220398 created "2020-10-22" @default.
• W3093220398 creator A5014676276 @default.
• W3093220398 creator A5018838133 @default.
• W3093220398 creator A5052441498 @default.
• W3093220398 creator A5079245916 @default.
• W3093220398 creator A5085245110 @default.
• W3093220398 date "2020-10-12" @default.
• W3093220398 modified "2023-10-18" @default.
• W3093220398 title "Forest R-CNN: Large-Vocabulary Long-Tailed Object Detection and Instance Segmentation" @default.
• W3093220398 cites W1536680647 @default.
• W3093220398 cites W1861492603 @default.
• W3093220398 cites W2031489346 @default.
• W3093220398 cites W2102605133 @default.
• W3093220398 cites W2132791018 @default.
• W3093220398 cites W2194775991 @default.
• W3093220398 cites W2317851288 @default.
• W3093220398 cites W2339172597 @default.
• W3093220398 cites W2440599146 @default.
• W3093220398 cites W2469885745 @default.
• W3093220398 cites W2555182955 @default.
• W3093220398 cites W2570343428 @default.
• W3093220398 cites W2781292787 @default.
• W3093220398 cites W2894299524 @default.
• W3093220398 cites W2920326761 @default.
• W3093220398 cites W2948672349 @default.
• W3093220398 cites W2963150697 @default.
• W3093220398 cites W2963351448 @default.
• W3093220398 cites W2963574614 @default.
• W3093220398 cites W2963691377 @default.
• W3093220398 cites W2963857746 @default.
• W3093220398 cites W2970084480 @default.
• W3093220398 cites W2985405845 @default.
• W3093220398 cites W2990361805 @default.
• W3093220398 cites W2993182889 @default.
• W3093220398 cites W3034601242 @default.
• W3093220398 cites W3034933032 @default.
• W3093220398 cites W3034934229 @default.
• W3093220398 cites W3034955056 @default.
• W3093220398 cites W3035552357 @default.
• W3093220398 cites W3092873533 @default.
• W3093220398 doi "https://doi.org/10.1145/3394171.3413970" @default.
• W3093220398 hasPublicationYear "2020" @default.
• W3093220398 type Work @default.
• W3093220398 sameAs 3093220398 @default.
• W3093220398 citedByCount "47" @default.
• W3093220398 countsByYear W30932203982020 @default.
• W3093220398 countsByYear W30932203982021 @default.
• W3093220398 countsByYear W30932203982022 @default.
• W3093220398 countsByYear W30932203982023 @default.
• W3093220398 crossrefType "proceedings-article" @default.
• W3093220398 hasAuthorship W3093220398A5014676276 @default.
• W3093220398 hasAuthorship W3093220398A5018838133 @default.
• W3093220398 hasAuthorship W3093220398A5052441498 @default.
• W3093220398 hasAuthorship W3093220398A5079245916 @default.
• W3093220398 hasAuthorship W3093220398A5085245110 @default.
• W3093220398 hasBestOaLocation W30932203982 @default.
• W3093220398 hasConcept C113174947 @default.
• W3093220398 hasConcept C115961682 @default.
• W3093220398 hasConcept C119857082 @default.
• W3093220398 hasConcept C134306372 @default.
• W3093220398 hasConcept C138885662 @default.
• W3093220398 hasConcept C153180895 @default.
• W3093220398 hasConcept C154945302 @default.
• W3093220398 hasConcept C199360897 @default.
• W3093220398 hasConcept C2776151529 @default.
• W3093220398 hasConcept C2777212361 @default.
• W3093220398 hasConcept C2777601683 @default.
• W3093220398 hasConcept C2780801425 @default.
• W3093220398 hasConcept C2781238097 @default.
• W3093220398 hasConcept C33923547 @default.
• W3093220398 hasConcept C41008148 @default.
• W3093220398 hasConcept C41895202 @default.
• W3093220398 hasConcept C75294576 @default.
• W3093220398 hasConcept C89600930 @default.
• W3093220398 hasConcept C94124525 @default.
• W3093220398 hasConcept C95623464 @default.
• W3093220398 hasConceptScore W3093220398C113174947 @default.
• W3093220398 hasConceptScore W3093220398C115961682 @default.
• W3093220398 hasConceptScore W3093220398C119857082 @default.
• W3093220398 hasConceptScore W3093220398C134306372 @default.
• W3093220398 hasConceptScore W3093220398C138885662 @default.
• W3093220398 hasConceptScore W3093220398C153180895 @default.
• W3093220398 hasConceptScore W3093220398C154945302 @default.
• W3093220398 hasConceptScore W3093220398C199360897 @default.
• W3093220398 hasConceptScore W3093220398C2776151529 @default.
• W3093220398 hasConceptScore W3093220398C2777212361 @default.
• W3093220398 hasConceptScore W3093220398C2777601683 @default.
• W3093220398 hasConceptScore W3093220398C2780801425 @default.
• W3093220398 hasConceptScore W3093220398C2781238097 @default.
• W3093220398 hasConceptScore W3093220398C33923547 @default.
• W3093220398 hasConceptScore W3093220398C41008148 @default.
• W3093220398 hasConceptScore W3093220398C41895202 @default.
• W3093220398 hasConceptScore W3093220398C75294576 @default.
• W3093220398 hasConceptScore W3093220398C89600930 @default.
• W3093220398 hasConceptScore W3093220398C94124525 @default.
• W3093220398 hasConceptScore W3093220398C95623464 @default.
• W3093220398 hasLocation W30932203981 @default.
• W3093220398 hasLocation W30932203982 @default.
• W3093220398 hasOpenAccess W3093220398 @default.

• next