Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204163701> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W3204163701 endingPage "5127" @default.
- W3204163701 startingPage "5117" @default.
- W3204163701 abstract "Recently, deep Convolutional Neural Networks (CNNs) can achieve human-levelperformance in edge detection with the rich and abstract edge representationcapacities. However, the high performance of CNN based edge detection isachieved with a large pretrained CNN backbone, which is memory and energyconsuming. In addition, it is surprising that the previous wisdom from thetraditional edge detectors, such as Canny, Sobel, and LBP are rarelyinvestigated in the rapid-developing deep learning era. To address theseissues, we propose a simple, lightweight yet effective architecture named PixelDifference Network (PiDiNet) for efficient edge detection. Extensiveexperiments on BSDS500, NYUD, and Multicue are provided to demonstrate itseffectiveness, and its high training and inference efficiency. Surprisingly,when training from scratch with only the BSDS500 and VOC datasets, PiDiNet cansurpass the recorded result of human perception (0.807 vs. 0.803 in ODSF-measure) on the BSDS500 dataset with 100 FPS and less than 1M parameters. Afaster version of PiDiNet with less than 0.1M parameters can still achievecomparable performance among state of the arts with 200 FPS. Results on theNYUD and Multicue datasets show similar observations. The codes are availableat https://github.com/zhuoinoulu/pidinet." @default.
- W3204163701 created "2021-10-11" @default.
- W3204163701 creator A5010794708 @default.
- W3204163701 creator A5014611768 @default.
- W3204163701 creator A5039142337 @default.
- W3204163701 creator A5047455588 @default.
- W3204163701 creator A5062522283 @default.
- W3204163701 creator A5063481044 @default.
- W3204163701 creator A5069241620 @default.
- W3204163701 creator A5071074935 @default.
- W3204163701 date "2021-08-16" @default.
- W3204163701 modified "2023-09-25" @default.
- W3204163701 title "Pixel Difference Networks for Efficient Edge Detection" @default.
- W3204163701 hasPublicationYear "2021" @default.
- W3204163701 type Work @default.
- W3204163701 sameAs 3204163701 @default.
- W3204163701 citedByCount "2" @default.
- W3204163701 countsByYear W32041637012021 @default.
- W3204163701 crossrefType "proceedings-article" @default.
- W3204163701 hasAuthorship W3204163701A5010794708 @default.
- W3204163701 hasAuthorship W3204163701A5014611768 @default.
- W3204163701 hasAuthorship W3204163701A5039142337 @default.
- W3204163701 hasAuthorship W3204163701A5047455588 @default.
- W3204163701 hasAuthorship W3204163701A5062522283 @default.
- W3204163701 hasAuthorship W3204163701A5063481044 @default.
- W3204163701 hasAuthorship W3204163701A5069241620 @default.
- W3204163701 hasAuthorship W3204163701A5071074935 @default.
- W3204163701 hasConcept C108583219 @default.
- W3204163701 hasConcept C115961682 @default.
- W3204163701 hasConcept C14705441 @default.
- W3204163701 hasConcept C153180895 @default.
- W3204163701 hasConcept C154945302 @default.
- W3204163701 hasConcept C160633673 @default.
- W3204163701 hasConcept C162307627 @default.
- W3204163701 hasConcept C193536780 @default.
- W3204163701 hasConcept C2776214188 @default.
- W3204163701 hasConcept C30703548 @default.
- W3204163701 hasConcept C31972630 @default.
- W3204163701 hasConcept C41008148 @default.
- W3204163701 hasConcept C81363708 @default.
- W3204163701 hasConcept C9417928 @default.
- W3204163701 hasConceptScore W3204163701C108583219 @default.
- W3204163701 hasConceptScore W3204163701C115961682 @default.
- W3204163701 hasConceptScore W3204163701C14705441 @default.
- W3204163701 hasConceptScore W3204163701C153180895 @default.
- W3204163701 hasConceptScore W3204163701C154945302 @default.
- W3204163701 hasConceptScore W3204163701C160633673 @default.
- W3204163701 hasConceptScore W3204163701C162307627 @default.
- W3204163701 hasConceptScore W3204163701C193536780 @default.
- W3204163701 hasConceptScore W3204163701C2776214188 @default.
- W3204163701 hasConceptScore W3204163701C30703548 @default.
- W3204163701 hasConceptScore W3204163701C31972630 @default.
- W3204163701 hasConceptScore W3204163701C41008148 @default.
- W3204163701 hasConceptScore W3204163701C81363708 @default.
- W3204163701 hasConceptScore W3204163701C9417928 @default.
- W3204163701 hasLocation W32041637011 @default.
- W3204163701 hasOpenAccess W3204163701 @default.
- W3204163701 hasPrimaryLocation W32041637011 @default.
- W3204163701 hasRelatedWork W2267459590 @default.
- W3204163701 hasRelatedWork W2302502886 @default.
- W3204163701 hasRelatedWork W2736384776 @default.
- W3204163701 hasRelatedWork W2768704901 @default.
- W3204163701 hasRelatedWork W2773412982 @default.
- W3204163701 hasRelatedWork W2781409998 @default.
- W3204163701 hasRelatedWork W2807333821 @default.
- W3204163701 hasRelatedWork W2897840973 @default.
- W3204163701 hasRelatedWork W2908050327 @default.
- W3204163701 hasRelatedWork W2921870894 @default.
- W3204163701 hasRelatedWork W2955664919 @default.
- W3204163701 hasRelatedWork W2973081974 @default.
- W3204163701 hasRelatedWork W2995705876 @default.
- W3204163701 hasRelatedWork W3014700429 @default.
- W3204163701 hasRelatedWork W3016720554 @default.
- W3204163701 hasRelatedWork W3081621190 @default.
- W3204163701 hasRelatedWork W3120290381 @default.
- W3204163701 hasRelatedWork W3153190002 @default.
- W3204163701 hasRelatedWork W3194146936 @default.
- W3204163701 hasRelatedWork W3194258098 @default.
- W3204163701 isParatext "false" @default.
- W3204163701 isRetracted "false" @default.
- W3204163701 magId "3204163701" @default.
- W3204163701 workType "article" @default.