Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320900857> ?p ?o ?g. }
- W4320900857 endingPage "326" @default.
- W4320900857 startingPage "308" @default.
- W4320900857 abstract "With an unprecedented increase in the number of agents and systems that aim to navigate the real world using visual cues and the rising impetus for 3D Vision Models, the importance of depth estimation is hard to understate. While supervised methods remain the gold standard in the domain, the copious amount of paired stereo data required to train such models makes them impractical. Most State of the Art (SOTA) works in the self-supervised and unsupervised domain employ a ResNet-based encoder architecture to predict disparity maps from a given input image which are eventually used alongside a camera pose estimator to predict depth without direct supervision. The fully convolutional nature of ResNets makes them susceptible to capturing per-pixel local information only, which is suboptimal for depth prediction. Our key insight for doing away with this bottleneck is to use Vision Transformers, which employ self-attention to capture the global contextual information present in an input image. Our model fuses per-pixel local information learned using two fully convolutional depth encoders with global contextual information learned by a transformer encoder at different scales. It does so using a mask-guided multi-stream convolution in the feature space to achieve state-of-the-art performance on most standard benchmarks." @default.
- W4320900857 created "2023-02-16" @default.
- W4320900857 creator A5008386628 @default.
- W4320900857 creator A5072237625 @default.
- W4320900857 creator A5088829322 @default.
- W4320900857 date "2023-01-01" @default.
- W4320900857 modified "2023-10-01" @default.
- W4320900857 title "Hybrid Transformer Based Feature Fusion for Self-Supervised Monocular Depth Estimation" @default.
- W4320900857 cites W2132947399 @default.
- W4320900857 cites W2133665775 @default.
- W4320900857 cites W2150066425 @default.
- W4320900857 cites W2194775991 @default.
- W4320900857 cites W2476548250 @default.
- W4320900857 cites W2609883120 @default.
- W4320900857 cites W2798927139 @default.
- W4320900857 cites W2890949887 @default.
- W4320900857 cites W2948647700 @default.
- W4320900857 cites W2963265330 @default.
- W4320900857 cites W2963488291 @default.
- W4320900857 cites W2963906250 @default.
- W4320900857 cites W2964121744 @default.
- W4320900857 cites W2968529893 @default.
- W4320900857 cites W2981732213 @default.
- W4320900857 cites W2982102242 @default.
- W4320900857 cites W2983619469 @default.
- W4320900857 cites W2985775862 @default.
- W4320900857 cites W3014263713 @default.
- W4320900857 cites W3014641072 @default.
- W4320900857 cites W3034428934 @default.
- W4320900857 cites W3034604951 @default.
- W4320900857 cites W3034953156 @default.
- W4320900857 cites W3035563424 @default.
- W4320900857 cites W3035679448 @default.
- W4320900857 cites W3107156787 @default.
- W4320900857 cites W3117012253 @default.
- W4320900857 cites W3118942022 @default.
- W4320900857 cites W3121115395 @default.
- W4320900857 cites W3129840128 @default.
- W4320900857 cites W3136544170 @default.
- W4320900857 cites W3174541782 @default.
- W4320900857 cites W3176063055 @default.
- W4320900857 cites W3178217626 @default.
- W4320900857 cites W3188851454 @default.
- W4320900857 cites W3207918547 @default.
- W4320900857 cites W3215899623 @default.
- W4320900857 cites W4213295893 @default.
- W4320900857 cites W4214520160 @default.
- W4320900857 cites W4225672218 @default.
- W4320900857 cites W4226250168 @default.
- W4320900857 cites W4312238440 @default.
- W4320900857 cites W4312509967 @default.
- W4320900857 cites W4312546175 @default.
- W4320900857 cites W4312726009 @default.
- W4320900857 cites W4312812783 @default.
- W4320900857 cites W4312980231 @default.
- W4320900857 doi "https://doi.org/10.1007/978-3-031-25063-7_19" @default.
- W4320900857 hasPublicationYear "2023" @default.
- W4320900857 type Work @default.
- W4320900857 citedByCount "1" @default.
- W4320900857 countsByYear W43209008572023 @default.
- W4320900857 crossrefType "book-chapter" @default.
- W4320900857 hasAuthorship W4320900857A5008386628 @default.
- W4320900857 hasAuthorship W4320900857A5072237625 @default.
- W4320900857 hasAuthorship W4320900857A5088829322 @default.
- W4320900857 hasBestOaLocation W43209008572 @default.
- W4320900857 hasConcept C111919701 @default.
- W4320900857 hasConcept C118505674 @default.
- W4320900857 hasConcept C121332964 @default.
- W4320900857 hasConcept C149635348 @default.
- W4320900857 hasConcept C153180895 @default.
- W4320900857 hasConcept C154945302 @default.
- W4320900857 hasConcept C165801399 @default.
- W4320900857 hasConcept C2780513914 @default.
- W4320900857 hasConcept C31972630 @default.
- W4320900857 hasConcept C41008148 @default.
- W4320900857 hasConcept C62520636 @default.
- W4320900857 hasConcept C65909025 @default.
- W4320900857 hasConcept C66322947 @default.
- W4320900857 hasConcept C81363708 @default.
- W4320900857 hasConceptScore W4320900857C111919701 @default.
- W4320900857 hasConceptScore W4320900857C118505674 @default.
- W4320900857 hasConceptScore W4320900857C121332964 @default.
- W4320900857 hasConceptScore W4320900857C149635348 @default.
- W4320900857 hasConceptScore W4320900857C153180895 @default.
- W4320900857 hasConceptScore W4320900857C154945302 @default.
- W4320900857 hasConceptScore W4320900857C165801399 @default.
- W4320900857 hasConceptScore W4320900857C2780513914 @default.
- W4320900857 hasConceptScore W4320900857C31972630 @default.
- W4320900857 hasConceptScore W4320900857C41008148 @default.
- W4320900857 hasConceptScore W4320900857C62520636 @default.
- W4320900857 hasConceptScore W4320900857C65909025 @default.
- W4320900857 hasConceptScore W4320900857C66322947 @default.
- W4320900857 hasConceptScore W4320900857C81363708 @default.
- W4320900857 hasLocation W43209008571 @default.
- W4320900857 hasLocation W43209008572 @default.
- W4320900857 hasOpenAccess W4320900857 @default.
- W4320900857 hasPrimaryLocation W43209008571 @default.
- W4320900857 hasRelatedWork W127013308 @default.