FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3112615861> ?p ?o ?g. }

• W3112615861 abstract "Despite significant progress in monocular depth estimation in the wild, recent state-of-the-art methods cannot be used to recover accurate 3D scene shape due to an unknown depth shift induced by shift-invariant reconstruction losses used in mixed-data depth prediction training, and possible unknown camera focal length. We investigate this problem in detail, and propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image, and then use 3D point cloud encoders to predict the missing depth shift and focal length that allow us to recover a realistic 3D scene shape. In addition, we propose an image-level normalized regression loss and a normal-based geometry loss to enhance depth prediction models trained on mixed datasets. We test our depth model on nine unseen datasets and achieve state-of-the-art performance on zero-shot dataset generalization. Code is available at: https://git.io/Depth" @default.
• W3112615861 created "2020-12-21" @default.
• W3112615861 creator A5006294869 @default.
• W3112615861 creator A5010310339 @default.
• W3112615861 creator A5014702970 @default.
• W3112615861 creator A5022192060 @default.
• W3112615861 creator A5041338935 @default.
• W3112615861 creator A5067580558 @default.
• W3112615861 creator A5074776688 @default.
• W3112615861 date "2020-12-16" @default.
• W3112615861 modified "2023-10-16" @default.
• W3112615861 title "Learning to Recover 3D Scene Shape from a Single Image" @default.
• W3112615861 cites W125693051 @default.
• W3112615861 cites W1513100184 @default.
• W3112615861 cites W1576834227 @default.
• W3112615861 cites W1803059841 @default.
• W3112615861 cites W2027560260 @default.
• W3112615861 cites W2033819227 @default.
• W3112615861 cites W2103997188 @default.
• W3112615861 cites W2132947399 @default.
• W3112615861 cites W2135346934 @default.
• W3112615861 cites W2150066425 @default.
• W3112615861 cites W2160398734 @default.
• W3112615861 cites W2163634044 @default.
• W3112615861 cites W2167667767 @default.
• W3112615861 cites W2171740948 @default.
• W3112615861 cites W2194775991 @default.
• W3112615861 cites W2294109271 @default.
• W3112615861 cites W2549139847 @default.
• W3112615861 cites W2560474170 @default.
• W3112615861 cites W2586114507 @default.
• W3112615861 cites W2594519801 @default.
• W3112615861 cites W2741885505 @default.
• W3112615861 cites W2798373498 @default.
• W3112615861 cites W2803168974 @default.
• W3112615861 cites W2890382763 @default.
• W3112615861 cites W2945020349 @default.
• W3112615861 cites W2952813711 @default.
• W3112615861 cites W2962778872 @default.
• W3112615861 cites W2962833508 @default.
• W3112615861 cites W2963264757 @default.
• W3112615861 cites W2963760790 @default.
• W3112615861 cites W2964185501 @default.
• W3112615861 cites W2964309882 @default.
• W3112615861 cites W2965851327 @default.
• W3112615861 cites W2971499585 @default.
• W3112615861 cites W2981978060 @default.
• W3112615861 cites W2982336692 @default.
• W3112615861 cites W2985299701 @default.
• W3112615861 cites W2990946490 @default.
• W3112615861 cites W2998421254 @default.
• W3112615861 cites W3003892866 @default.
• W3112615861 cites W3013428964 @default.
• W3112615861 cites W3035291735 @default.
• W3112615861 cites W3035508487 @default.
• W3112615861 cites W3035563424 @default.
• W3112615861 cites W3035679448 @default.
• W3112615861 cites W3081167590 @default.
• W3112615861 cites W575847903 @default.
• W3112615861 doi "https://doi.org/10.48550/arxiv.2012.09365" @default.
• W3112615861 hasPublicationYear "2020" @default.
• W3112615861 type Work @default.
• W3112615861 sameAs 3112615861 @default.
• W3112615861 citedByCount "1" @default.
• W3112615861 countsByYear W31126158612023 @default.
• W3112615861 crossrefType "posted-content" @default.
• W3112615861 hasAuthorship W3112615861A5006294869 @default.
• W3112615861 hasAuthorship W3112615861A5010310339 @default.
• W3112615861 hasAuthorship W3112615861A5014702970 @default.
• W3112615861 hasAuthorship W3112615861A5022192060 @default.
• W3112615861 hasAuthorship W3112615861A5041338935 @default.
• W3112615861 hasAuthorship W3112615861A5067580558 @default.
• W3112615861 hasAuthorship W3112615861A5074776688 @default.
• W3112615861 hasBestOaLocation W31126158611 @default.
• W3112615861 hasConcept C111919701 @default.
• W3112615861 hasConcept C115961682 @default.
• W3112615861 hasConcept C118505674 @default.
• W3112615861 hasConcept C131979681 @default.
• W3112615861 hasConcept C154945302 @default.
• W3112615861 hasConcept C177264268 @default.
• W3112615861 hasConcept C190470478 @default.
• W3112615861 hasConcept C199360897 @default.
• W3112615861 hasConcept C2776760102 @default.
• W3112615861 hasConcept C31972630 @default.
• W3112615861 hasConcept C33923547 @default.
• W3112615861 hasConcept C37914503 @default.
• W3112615861 hasConcept C41008148 @default.
• W3112615861 hasConcept C65909025 @default.
• W3112615861 hasConceptScore W3112615861C111919701 @default.
• W3112615861 hasConceptScore W3112615861C115961682 @default.
• W3112615861 hasConceptScore W3112615861C118505674 @default.
• W3112615861 hasConceptScore W3112615861C131979681 @default.
• W3112615861 hasConceptScore W3112615861C154945302 @default.
• W3112615861 hasConceptScore W3112615861C177264268 @default.
• W3112615861 hasConceptScore W3112615861C190470478 @default.
• W3112615861 hasConceptScore W3112615861C199360897 @default.
• W3112615861 hasConceptScore W3112615861C2776760102 @default.
• W3112615861 hasConceptScore W3112615861C31972630 @default.
• W3112615861 hasConceptScore W3112615861C33923547 @default.
• W3112615861 hasConceptScore W3112615861C37914503 @default.

• next