FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308173664> ?p ?o ?g. }

• W4308173664 endingPage "5347" @default.
• W4308173664 startingPage "5347" @default.
• W4308173664 abstract "In recent years, object detectors based on convolutional neural networks have been widely used on remote sensing images. However, the improvement of their detection performance depends on a deeper convolution layer and a complex convolution structure, resulting in a significant increase in the storage space and computational complexity. Although previous works have designed a variety of new lightweight convolution and compression algorithms, these works often require complex manual design and cause the detector to be greatly modified, which makes it difficult to directly apply the algorithms to different detectors and general hardware. Therefore, this paper proposes an iterative pruning framework based on assistant distillation. Specifically, a structured sparse pruning strategy for detectors is proposed. By taking the channel scaling factor as a representation of the weight importance, the channels of the network are pruned and the detector is greatly slimmed. Then, a teacher assistant distillation model is proposed to recover the network performance after compression. The intermediate models retained in the pruning process are used as assistant models. By way of the teachers distilling the assistants and the assistants distilling the students, the students’ underfitting caused by the difference in capacity between teachers and students is eliminated, thus effectively restoring the network performance. By using this compression framework, we can greatly compress the network without changing the network structure and can obtain the support of any hardware platform and deep learning library. Extensive experiments show that compared with existing detection networks, our method can achieve an effective balance between speed and accuracy on three commonly used remote sensing target datasets (i.e., NWPU VHR-10, RSOD, and DOTA)." @default.
• W4308173664 created "2022-11-08" @default.
• W4308173664 creator A5024498727 @default.
• W4308173664 creator A5044435891 @default.
• W4308173664 creator A5079668307 @default.
• W4308173664 creator A5090017494 @default.
• W4308173664 date "2022-10-25" @default.
• W4308173664 modified "2023-10-11" @default.
• W4308173664 title "Sparse Channel Pruning and Assistant Distillation for Faster Aerial Object Detection" @default.
• W4308173664 cites W2061200049 @default.
• W4308173664 cites W2294370754 @default.
• W4308173664 cites W2596246567 @default.
• W4308173664 cites W2743289088 @default.
• W4308173664 cites W2792075695 @default.
• W4308173664 cites W2903707108 @default.
• W4308173664 cites W2963037989 @default.
• W4308173664 cites W2989604896 @default.
• W4308173664 cites W3004127093 @default.
• W4308173664 cites W3101012758 @default.
• W4308173664 cites W3106228955 @default.
• W4308173664 cites W3106250896 @default.
• W4308173664 cites W3136761610 @default.
• W4308173664 cites W3174873843 @default.
• W4308173664 cites W3214821343 @default.
• W4308173664 cites W4212908285 @default.
• W4308173664 cites W4214535086 @default.
• W4308173664 cites W4226034202 @default.
• W4308173664 cites W4229367296 @default.
• W4308173664 cites W639708223 @default.
• W4308173664 doi "https://doi.org/10.3390/rs14215347" @default.
• W4308173664 hasPublicationYear "2022" @default.
• W4308173664 type Work @default.
• W4308173664 citedByCount "2" @default.
• W4308173664 countsByYear W43081736642023 @default.
• W4308173664 crossrefType "journal-article" @default.
• W4308173664 hasAuthorship W4308173664A5024498727 @default.
• W4308173664 hasAuthorship W4308173664A5044435891 @default.
• W4308173664 hasAuthorship W4308173664A5079668307 @default.
• W4308173664 hasAuthorship W4308173664A5090017494 @default.
• W4308173664 hasBestOaLocation W43081736641 @default.
• W4308173664 hasConcept C108010975 @default.
• W4308173664 hasConcept C108583219 @default.
• W4308173664 hasConcept C111919701 @default.
• W4308173664 hasConcept C113775141 @default.
• W4308173664 hasConcept C11413529 @default.
• W4308173664 hasConcept C119857082 @default.
• W4308173664 hasConcept C127162648 @default.
• W4308173664 hasConcept C153180895 @default.
• W4308173664 hasConcept C154945302 @default.
• W4308173664 hasConcept C17744445 @default.
• W4308173664 hasConcept C199539241 @default.
• W4308173664 hasConcept C2776151529 @default.
• W4308173664 hasConcept C2776359362 @default.
• W4308173664 hasConcept C31258907 @default.
• W4308173664 hasConcept C41008148 @default.
• W4308173664 hasConcept C45347329 @default.
• W4308173664 hasConcept C50644808 @default.
• W4308173664 hasConcept C6557445 @default.
• W4308173664 hasConcept C76155785 @default.
• W4308173664 hasConcept C81363708 @default.
• W4308173664 hasConcept C86803240 @default.
• W4308173664 hasConcept C94625758 @default.
• W4308173664 hasConcept C94915269 @default.
• W4308173664 hasConcept C98045186 @default.
• W4308173664 hasConceptScore W4308173664C108010975 @default.
• W4308173664 hasConceptScore W4308173664C108583219 @default.
• W4308173664 hasConceptScore W4308173664C111919701 @default.
• W4308173664 hasConceptScore W4308173664C113775141 @default.
• W4308173664 hasConceptScore W4308173664C11413529 @default.
• W4308173664 hasConceptScore W4308173664C119857082 @default.
• W4308173664 hasConceptScore W4308173664C127162648 @default.
• W4308173664 hasConceptScore W4308173664C153180895 @default.
• W4308173664 hasConceptScore W4308173664C154945302 @default.
• W4308173664 hasConceptScore W4308173664C17744445 @default.
• W4308173664 hasConceptScore W4308173664C199539241 @default.
• W4308173664 hasConceptScore W4308173664C2776151529 @default.
• W4308173664 hasConceptScore W4308173664C2776359362 @default.
• W4308173664 hasConceptScore W4308173664C31258907 @default.
• W4308173664 hasConceptScore W4308173664C41008148 @default.
• W4308173664 hasConceptScore W4308173664C45347329 @default.
• W4308173664 hasConceptScore W4308173664C50644808 @default.
• W4308173664 hasConceptScore W4308173664C6557445 @default.
• W4308173664 hasConceptScore W4308173664C76155785 @default.
• W4308173664 hasConceptScore W4308173664C81363708 @default.
• W4308173664 hasConceptScore W4308173664C86803240 @default.
• W4308173664 hasConceptScore W4308173664C94625758 @default.
• W4308173664 hasConceptScore W4308173664C94915269 @default.
• W4308173664 hasConceptScore W4308173664C98045186 @default.
• W4308173664 hasFunder F4320321001 @default.
• W4308173664 hasFunder F4320321543 @default.
• W4308173664 hasIssue "21" @default.
• W4308173664 hasLocation W43081736641 @default.
• W4308173664 hasLocation W43081736642 @default.
• W4308173664 hasOpenAccess W4308173664 @default.
• W4308173664 hasPrimaryLocation W43081736641 @default.
• W4308173664 hasRelatedWork W2731899572 @default.
• W4308173664 hasRelatedWork W2999805992 @default.
• W4308173664 hasRelatedWork W3116150086 @default.

• next