FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3173561529> ?p ?o ?g. }

• W3173561529 abstract "Estimating 3D scene flow from a sequence of monocular images has been gaining increased attention due to the simple, economical capture setup. Owing to the severe ill-posedness of the problem, the accuracy of current methods has been limited, especially that of efficient, real-time approaches. In this paper, we introduce a multi-frame monocular scene flow network based on self-supervised learning, improving the accuracy over previous networks while retaining real-time efficiency. Based on an advanced two-frame baseline with a split-decoder design, we propose (i) a multi-frame model using a triple frame input and convolutional LSTM connections, (ii) an occlusion-aware census loss for better accuracy, and (iii) a gradient detaching strategy to improve training stability. On the KITTI dataset, we observe state-of-the-art accuracy among monocular scene flow methods based on self-supervised learning." @default.
• W3173561529 created "2021-07-05" @default.
• W3173561529 creator A5005363574 @default.
• W3173561529 creator A5056098625 @default.
• W3173561529 date "2021-06-01" @default.
• W3173561529 modified "2023-10-03" @default.
• W3173561529 title "Self-Supervised Multi-Frame Monocular Scene Flow" @default.
• W3173561529 cites W119960974 @default.
• W3173561529 cites W1499225944 @default.
• W3173561529 cites W179654571 @default.
• W3173561529 cites W181676623 @default.
• W3173561529 cites W1978881539 @default.
• W3173561529 cites W1993850091 @default.
• W3173561529 cites W2024336175 @default.
• W3173561529 cites W2064675550 @default.
• W3173561529 cites W2078113769 @default.
• W3173561529 cites W2087921381 @default.
• W3173561529 cites W2088616581 @default.
• W3173561529 cites W2088692258 @default.
• W3173561529 cites W2098500213 @default.
• W3173561529 cites W2115579991 @default.
• W3173561529 cites W2126184431 @default.
• W3173561529 cites W2129671742 @default.
• W3173561529 cites W2131952444 @default.
• W3173561529 cites W2157719643 @default.
• W3173561529 cites W2168538937 @default.
• W3173561529 cites W2286655030 @default.
• W3173561529 cites W2470139095 @default.
• W3173561529 cites W2520707372 @default.
• W3173561529 cites W2737529999 @default.
• W3173561529 cites W2770902190 @default.
• W3173561529 cites W2779333428 @default.
• W3173561529 cites W2887479417 @default.
• W3173561529 cites W2904340070 @default.
• W3173561529 cites W2949023359 @default.
• W3173561529 cites W2953804911 @default.
• W3173561529 cites W2962771259 @default.
• W3173561529 cites W2962801140 @default.
• W3173561529 cites W2962978889 @default.
• W3173561529 cites W2962989121 @default.
• W3173561529 cites W2963265330 @default.
• W3173561529 cites W2963583471 @default.
• W3173561529 cites W2963652593 @default.
• W3173561529 cites W2963734691 @default.
• W3173561529 cites W2963782415 @default.
• W3173561529 cites W2963954891 @default.
• W3173561529 cites W2964677288 @default.
• W3173561529 cites W2964968086 @default.
• W3173561529 cites W2965770986 @default.
• W3173561529 cites W2967043539 @default.
• W3173561529 cites W2970881233 @default.
• W3173561529 cites W2971686478 @default.
• W3173561529 cites W2981732213 @default.
• W3173561529 cites W2983921198 @default.
• W3173561529 cites W2984858187 @default.
• W3173561529 cites W2985178537 @default.
• W3173561529 cites W2985775862 @default.
• W3173561529 cites W2998488573 @default.
• W3173561529 cites W3003805155 @default.
• W3173561529 cites W3004309688 @default.
• W3173561529 cites W3010090948 @default.
• W3173561529 cites W3034193257 @default.
• W3173561529 cites W3034321406 @default.
• W3173561529 cites W3034642839 @default.
• W3173561529 cites W3034921716 @default.
• W3173561529 cites W3035574168 @default.
• W3173561529 cites W3098068165 @default.
• W3173561529 cites W3100388886 @default.
• W3173561529 cites W988642845 @default.
• W3173561529 doi "https://doi.org/10.1109/cvpr46437.2021.00271" @default.
• W3173561529 hasPublicationYear "2021" @default.
• W3173561529 type Work @default.
• W3173561529 sameAs 3173561529 @default.
• W3173561529 citedByCount "23" @default.
• W3173561529 countsByYear W31735615292021 @default.
• W3173561529 countsByYear W31735615292022 @default.
• W3173561529 countsByYear W31735615292023 @default.
• W3173561529 crossrefType "proceedings-article" @default.
• W3173561529 hasAuthorship W3173561529A5005363574 @default.
• W3173561529 hasAuthorship W3173561529A5056098625 @default.
• W3173561529 hasBestOaLocation W31735615292 @default.
• W3173561529 hasConcept C108583219 @default.
• W3173561529 hasConcept C112972136 @default.
• W3173561529 hasConcept C115961682 @default.
• W3173561529 hasConcept C119857082 @default.
• W3173561529 hasConcept C126042441 @default.
• W3173561529 hasConcept C154945302 @default.
• W3173561529 hasConcept C155542232 @default.
• W3173561529 hasConcept C31972630 @default.
• W3173561529 hasConcept C41008148 @default.
• W3173561529 hasConcept C65909025 @default.
• W3173561529 hasConcept C76155785 @default.
• W3173561529 hasConceptScore W3173561529C108583219 @default.
• W3173561529 hasConceptScore W3173561529C112972136 @default.
• W3173561529 hasConceptScore W3173561529C115961682 @default.
• W3173561529 hasConceptScore W3173561529C119857082 @default.
• W3173561529 hasConceptScore W3173561529C126042441 @default.
• W3173561529 hasConceptScore W3173561529C154945302 @default.
• W3173561529 hasConceptScore W3173561529C155542232 @default.
• W3173561529 hasConceptScore W3173561529C31972630 @default.

• next