FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2903392100> ?p ?o ?g. }

• W2903392100 abstract "Optical flow estimation is a challenging task in computer vision. Recent methods formulate such task as a supervised-learning problem. But it often suffers from limited realistic ground truth. In this paper, a compact network, embedded with cost volume, residual encoder and deconvolutional decoder, is presented to regress optical flow in an end- to-end manner. To overcome the lack of flow labels, we propose a novel data-driven strategy called masked label learning, where a large amount of masked labels are generated from the FlowNet 2.0 model and filtered by warping calibration for model training. We also present an extended-Huber loss to handle large displacements. With pretraining on massive masked flow data, followed by finetuning on a small number of sparse labels, our method achieves state-of-the-art accuracy on KITTI flow benchmark." @default.
• W2903392100 created "2018-12-11" @default.
• W2903392100 creator A5013311767 @default.
• W2903392100 creator A5048226673 @default.
• W2903392100 creator A5056919856 @default.
• W2903392100 creator A5089499642 @default.
• W2903392100 date "2018-08-01" @default.
• W2903392100 modified "2023-09-27" @default.
• W2903392100 title "Masked Label Learning for Optical Flow Regression" @default.
• W2903392100 cites W1578285471 @default.
• W2903392100 cites W1903029394 @default.
• W2903392100 cites W1921093919 @default.
• W2903392100 cites W1951289974 @default.
• W2903392100 cites W1977194745 @default.
• W2903392100 cites W2010719245 @default.
• W2903392100 cites W2033959528 @default.
• W2903392100 cites W2100315781 @default.
• W2903392100 cites W2115606256 @default.
• W2903392100 cites W2131747574 @default.
• W2903392100 cites W2150066425 @default.
• W2903392100 cites W2155893237 @default.
• W2903392100 cites W2194775991 @default.
• W2903392100 cites W2294238219 @default.
• W2903392100 cites W2340897893 @default.
• W2903392100 cites W2548527721 @default.
• W2903392100 cites W2552900565 @default.
• W2903392100 cites W2558923462 @default.
• W2903392100 cites W2560474170 @default.
• W2903392100 cites W2738028674 @default.
• W2903392100 cites W2779333428 @default.
• W2903392100 cites W2963210183 @default.
• W2903392100 cites W2963219046 @default.
• W2903392100 cites W2963305757 @default.
• W2903392100 cites W2963317244 @default.
• W2903392100 cites W2963891416 @default.
• W2903392100 cites W2964140947 @default.
• W2903392100 cites W3100388886 @default.
• W2903392100 cites W764651262 @default.
• W2903392100 doi "https://doi.org/10.1109/icpr.2018.8545747" @default.
• W2903392100 hasPublicationYear "2018" @default.
• W2903392100 type Work @default.
• W2903392100 sameAs 2903392100 @default.
• W2903392100 citedByCount "0" @default.
• W2903392100 crossrefType "proceedings-article" @default.
• W2903392100 hasAuthorship W2903392100A5013311767 @default.
• W2903392100 hasAuthorship W2903392100A5048226673 @default.
• W2903392100 hasAuthorship W2903392100A5056919856 @default.
• W2903392100 hasAuthorship W2903392100A5089499642 @default.
• W2903392100 hasConcept C105795698 @default.
• W2903392100 hasConcept C108583219 @default.
• W2903392100 hasConcept C111919701 @default.
• W2903392100 hasConcept C11413529 @default.
• W2903392100 hasConcept C115961682 @default.
• W2903392100 hasConcept C118505674 @default.
• W2903392100 hasConcept C119857082 @default.
• W2903392100 hasConcept C127413603 @default.
• W2903392100 hasConcept C13280743 @default.
• W2903392100 hasConcept C146849305 @default.
• W2903392100 hasConcept C153180895 @default.
• W2903392100 hasConcept C154945302 @default.
• W2903392100 hasConcept C155512373 @default.
• W2903392100 hasConcept C155542232 @default.
• W2903392100 hasConcept C157202957 @default.
• W2903392100 hasConcept C165838908 @default.
• W2903392100 hasConcept C185798385 @default.
• W2903392100 hasConcept C201995342 @default.
• W2903392100 hasConcept C205649164 @default.
• W2903392100 hasConcept C2524010 @default.
• W2903392100 hasConcept C2780451532 @default.
• W2903392100 hasConcept C31972630 @default.
• W2903392100 hasConcept C33923547 @default.
• W2903392100 hasConcept C38349280 @default.
• W2903392100 hasConcept C41008148 @default.
• W2903392100 hasConceptScore W2903392100C105795698 @default.
• W2903392100 hasConceptScore W2903392100C108583219 @default.
• W2903392100 hasConceptScore W2903392100C111919701 @default.
• W2903392100 hasConceptScore W2903392100C11413529 @default.
• W2903392100 hasConceptScore W2903392100C115961682 @default.
• W2903392100 hasConceptScore W2903392100C118505674 @default.
• W2903392100 hasConceptScore W2903392100C119857082 @default.
• W2903392100 hasConceptScore W2903392100C127413603 @default.
• W2903392100 hasConceptScore W2903392100C13280743 @default.
• W2903392100 hasConceptScore W2903392100C146849305 @default.
• W2903392100 hasConceptScore W2903392100C153180895 @default.
• W2903392100 hasConceptScore W2903392100C154945302 @default.
• W2903392100 hasConceptScore W2903392100C155512373 @default.
• W2903392100 hasConceptScore W2903392100C155542232 @default.
• W2903392100 hasConceptScore W2903392100C157202957 @default.
• W2903392100 hasConceptScore W2903392100C165838908 @default.
• W2903392100 hasConceptScore W2903392100C185798385 @default.
• W2903392100 hasConceptScore W2903392100C201995342 @default.
• W2903392100 hasConceptScore W2903392100C205649164 @default.
• W2903392100 hasConceptScore W2903392100C2524010 @default.
• W2903392100 hasConceptScore W2903392100C2780451532 @default.
• W2903392100 hasConceptScore W2903392100C31972630 @default.
• W2903392100 hasConceptScore W2903392100C33923547 @default.
• W2903392100 hasConceptScore W2903392100C38349280 @default.
• W2903392100 hasConceptScore W2903392100C41008148 @default.
• W2903392100 hasLocation W29033921001 @default.
• W2903392100 hasOpenAccess W2903392100 @default.

• next