FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3036812908> ?p ?o ?g. }

• W3036812908 abstract "Based on the framework of multiple instance learning (MIL), tremendous works have promoted the advances of weakly supervised object detection (WSOD). However, most MIL-based methods tend to localize instances to their discriminative parts instead of the whole content. In this paper, we propose a spatial likelihood voting (SLV) module to converge the proposal localizing process without any bounding box annotations. Specifically, all region proposals in a given image play the role of voters every iteration during training, voting for the likelihood of each category in spatial dimensions. After dilating alignment on the area with large likelihood values, the voting results are regularized as bounding boxes, being used for the final classification and localization. Based on SLV, we further propose an end-to-end training framework for multi-task learning. The classification and localization tasks promote each other, which further improves the detection performance. Extensive experiments on the PASCAL VOC 2007 and 2012 datasets demonstrate the superior performance of SLV." @default.
• W3036812908 created "2020-06-25" @default.
• W3036812908 creator A5008526350 @default.
• W3036812908 creator A5024965898 @default.
• W3036812908 creator A5028024624 @default.
• W3036812908 creator A5031875473 @default.
• W3036812908 creator A5048905896 @default.
• W3036812908 date "2020-06-23" @default.
• W3036812908 modified "2023-10-11" @default.
• W3036812908 title "SLV: Spatial Likelihood Voting for Weakly Supervised Object Detection" @default.
• W3036812908 cites W1536680647 @default.
• W3036812908 cites W1686810756 @default.
• W3036812908 cites W1861492603 @default.
• W3036812908 cites W1934621328 @default.
• W3036812908 cites W2037227137 @default.
• W3036812908 cites W2088049833 @default.
• W3036812908 cites W2102605133 @default.
• W3036812908 cites W2112796928 @default.
• W3036812908 cites W2117539524 @default.
• W3036812908 cites W2133324800 @default.
• W3036812908 cites W2194775991 @default.
• W3036812908 cites W2248800885 @default.
• W3036812908 cites W2441255125 @default.
• W3036812908 cites W2519284461 @default.
• W3036812908 cites W2559348937 @default.
• W3036812908 cites W2565639579 @default.
• W3036812908 cites W2604260814 @default.
• W3036812908 cites W2610510511 @default.
• W3036812908 cites W2798269247 @default.
• W3036812908 cites W2798748179 @default.
• W3036812908 cites W2813911573 @default.
• W3036812908 cites W2884195989 @default.
• W3036812908 cites W2895236117 @default.
• W3036812908 cites W2899771611 @default.
• W3036812908 cites W2926593466 @default.
• W3036812908 cites W2949548022 @default.
• W3036812908 cites W2953106684 @default.
• W3036812908 cites W2963603913 @default.
• W3036812908 cites W2963949812 @default.
• W3036812908 cites W2963952323 @default.
• W3036812908 cites W2964328846 @default.
• W3036812908 cites W2990400263 @default.
• W3036812908 cites W2991023920 @default.
• W3036812908 cites W3106250896 @default.
• W3036812908 doi "https://doi.org/10.48550/arxiv.2006.12884" @default.
• W3036812908 hasPublicationYear "2020" @default.
• W3036812908 type Work @default.
• W3036812908 sameAs 3036812908 @default.
• W3036812908 citedByCount "0" @default.
• W3036812908 crossrefType "posted-content" @default.
• W3036812908 hasAuthorship W3036812908A5008526350 @default.
• W3036812908 hasAuthorship W3036812908A5024965898 @default.
• W3036812908 hasAuthorship W3036812908A5028024624 @default.
• W3036812908 hasAuthorship W3036812908A5031875473 @default.
• W3036812908 hasAuthorship W3036812908A5048905896 @default.
• W3036812908 hasBestOaLocation W30368129081 @default.
• W3036812908 hasConcept C111919701 @default.
• W3036812908 hasConcept C115961682 @default.
• W3036812908 hasConcept C119857082 @default.
• W3036812908 hasConcept C147037132 @default.
• W3036812908 hasConcept C153180895 @default.
• W3036812908 hasConcept C154945302 @default.
• W3036812908 hasConcept C162324750 @default.
• W3036812908 hasConcept C17744445 @default.
• W3036812908 hasConcept C187736073 @default.
• W3036812908 hasConcept C199360897 @default.
• W3036812908 hasConcept C199539241 @default.
• W3036812908 hasConcept C2776151529 @default.
• W3036812908 hasConcept C2780451532 @default.
• W3036812908 hasConcept C2781238097 @default.
• W3036812908 hasConcept C41008148 @default.
• W3036812908 hasConcept C520049643 @default.
• W3036812908 hasConcept C63584917 @default.
• W3036812908 hasConcept C75294576 @default.
• W3036812908 hasConcept C75608658 @default.
• W3036812908 hasConcept C94625758 @default.
• W3036812908 hasConcept C97931131 @default.
• W3036812908 hasConcept C98045186 @default.
• W3036812908 hasConceptScore W3036812908C111919701 @default.
• W3036812908 hasConceptScore W3036812908C115961682 @default.
• W3036812908 hasConceptScore W3036812908C119857082 @default.
• W3036812908 hasConceptScore W3036812908C147037132 @default.
• W3036812908 hasConceptScore W3036812908C153180895 @default.
• W3036812908 hasConceptScore W3036812908C154945302 @default.
• W3036812908 hasConceptScore W3036812908C162324750 @default.
• W3036812908 hasConceptScore W3036812908C17744445 @default.
• W3036812908 hasConceptScore W3036812908C187736073 @default.
• W3036812908 hasConceptScore W3036812908C199360897 @default.
• W3036812908 hasConceptScore W3036812908C199539241 @default.
• W3036812908 hasConceptScore W3036812908C2776151529 @default.
• W3036812908 hasConceptScore W3036812908C2780451532 @default.
• W3036812908 hasConceptScore W3036812908C2781238097 @default.
• W3036812908 hasConceptScore W3036812908C41008148 @default.
• W3036812908 hasConceptScore W3036812908C520049643 @default.
• W3036812908 hasConceptScore W3036812908C63584917 @default.
• W3036812908 hasConceptScore W3036812908C75294576 @default.
• W3036812908 hasConceptScore W3036812908C75608658 @default.
• W3036812908 hasConceptScore W3036812908C94625758 @default.
• W3036812908 hasConceptScore W3036812908C97931131 @default.
• W3036812908 hasConceptScore W3036812908C98045186 @default.

• next