Matches in SemOpenAlex for { <https://semopenalex.org/work/W3034157132> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W3034157132 abstract "In this paper, we present our solution for tiger detection in the 2019 Computer Vision for Wildlife Conservation Challenge (CVWC2019). We introduce an efficient deep tiger detector, which consists of the convnet channel adaptation method and an improved tiger detection method based on You Only Look Once version 3 (YOLOv3). Considering the limited memory and computing power of tiny embedded devices, we have used EfficientNet-B0 and Darknet-53 as backbone networks for detection and adapted them to balance their depth and width inspired by the channel pruning method and knowledge distillation method. Our results show that after an architecture adjustment of Darknet-53, the floating-point computation decreases by 93%, its model size decreases by 97%, and its accuracy only decreases by 1%; after an architecture adjustment of EfficientNet-B0, the floating-point computation decreases by 66%, its model size decreases by 70% with its accuracy only decreased by 1%. We also compare GIoU loss and MSE loss in the training stage. The GIoU loss has the advantage that it increases the average AP for IoU from 0.5 to 0.95 without affecting training speed and the interface speed, so it is experimentally reasonable for tiger detection in the wild. This proposed method outperforms previous Amur tiger detection methods presented at CVWC2019." @default.
- W3034157132 created "2020-06-12" @default.
- W3034157132 creator A5000410556 @default.
- W3034157132 creator A5005795572 @default.
- W3034157132 creator A5072833759 @default.
- W3034157132 date "2020-06-08" @default.
- W3034157132 modified "2023-10-14" @default.
- W3034157132 title "YOLO-mini-tiger: Amur Tiger Detection" @default.
- W3034157132 cites W2109255472 @default.
- W3034157132 cites W2194775991 @default.
- W3034157132 cites W2565639579 @default.
- W3034157132 cites W2570343428 @default.
- W3034157132 cites W2582323655 @default.
- W3034157132 cites W2747909401 @default.
- W3034157132 cites W2752782242 @default.
- W3034157132 cites W2962766617 @default.
- W3034157132 cites W2962851801 @default.
- W3034157132 cites W2963037989 @default.
- W3034157132 cites W2963163009 @default.
- W3034157132 cites W2963346150 @default.
- W3034157132 cites W2963446712 @default.
- W3034157132 cites W2963813458 @default.
- W3034157132 doi "https://doi.org/10.1145/3372278.3390710" @default.
- W3034157132 hasPublicationYear "2020" @default.
- W3034157132 type Work @default.
- W3034157132 sameAs 3034157132 @default.
- W3034157132 citedByCount "1" @default.
- W3034157132 countsByYear W30341571322023 @default.
- W3034157132 crossrefType "proceedings-article" @default.
- W3034157132 hasAuthorship W3034157132A5000410556 @default.
- W3034157132 hasAuthorship W3034157132A5005795572 @default.
- W3034157132 hasAuthorship W3034157132A5072833759 @default.
- W3034157132 hasConcept C108010975 @default.
- W3034157132 hasConcept C108583219 @default.
- W3034157132 hasConcept C11413529 @default.
- W3034157132 hasConcept C127162648 @default.
- W3034157132 hasConcept C148524875 @default.
- W3034157132 hasConcept C154945302 @default.
- W3034157132 hasConcept C2944601119 @default.
- W3034157132 hasConcept C41008148 @default.
- W3034157132 hasConcept C45374587 @default.
- W3034157132 hasConcept C6557445 @default.
- W3034157132 hasConcept C76155785 @default.
- W3034157132 hasConcept C79403827 @default.
- W3034157132 hasConcept C86803240 @default.
- W3034157132 hasConcept C91451191 @default.
- W3034157132 hasConcept C94915269 @default.
- W3034157132 hasConceptScore W3034157132C108010975 @default.
- W3034157132 hasConceptScore W3034157132C108583219 @default.
- W3034157132 hasConceptScore W3034157132C11413529 @default.
- W3034157132 hasConceptScore W3034157132C127162648 @default.
- W3034157132 hasConceptScore W3034157132C148524875 @default.
- W3034157132 hasConceptScore W3034157132C154945302 @default.
- W3034157132 hasConceptScore W3034157132C2944601119 @default.
- W3034157132 hasConceptScore W3034157132C41008148 @default.
- W3034157132 hasConceptScore W3034157132C45374587 @default.
- W3034157132 hasConceptScore W3034157132C6557445 @default.
- W3034157132 hasConceptScore W3034157132C76155785 @default.
- W3034157132 hasConceptScore W3034157132C79403827 @default.
- W3034157132 hasConceptScore W3034157132C86803240 @default.
- W3034157132 hasConceptScore W3034157132C91451191 @default.
- W3034157132 hasConceptScore W3034157132C94915269 @default.
- W3034157132 hasLocation W30341571321 @default.
- W3034157132 hasOpenAccess W3034157132 @default.
- W3034157132 hasPrimaryLocation W30341571321 @default.
- W3034157132 hasRelatedWork W1777228944 @default.
- W3034157132 hasRelatedWork W2246655820 @default.
- W3034157132 hasRelatedWork W2283566838 @default.
- W3034157132 hasRelatedWork W2799384463 @default.
- W3034157132 hasRelatedWork W3081748421 @default.
- W3034157132 hasRelatedWork W3159417908 @default.
- W3034157132 hasRelatedWork W4285161415 @default.
- W3034157132 hasRelatedWork W4383426653 @default.
- W3034157132 hasRelatedWork W4384699178 @default.
- W3034157132 hasRelatedWork W607825664 @default.
- W3034157132 isParatext "false" @default.
- W3034157132 isRetracted "false" @default.
- W3034157132 magId "3034157132" @default.
- W3034157132 workType "article" @default.