Matches in SemOpenAlex for { <https://semopenalex.org/work/W4315704975> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4315704975 abstract "Monocular depth estimation has been widely studied, and significant improvements in performance have been recently reported. However, most previous works are evaluated on a few benchmark datasets, such as KITTI datasets, and none of the works provide an in-depth analysis of the generalization performance of monocular depth estimation. In this paper, we deeply investigate the various backbone networks (e.g.CNN and Transformer models) toward the generalization of monocular depth estimation. First, we evaluate state-of-the-art models on both in-distribution and out-of-distribution datasets, which have never been seen during network training. Then, we investigate the internal properties of the representations from the intermediate layers of CNN-/Transformer-based models using synthetic texture-shifted datasets. Through extensive experiments, we observe that the Transformers exhibit a strong shape-bias rather than CNNs, which have a strong texture-bias. We also discover that texture-biased models exhibit worse generalization performance for monocular depth estimation than shape-biased models. We demonstrate that similar aspects are observed in real-world driving datasets captured under diverse environments. Lastly, we conduct a dense ablation study with various backbone networks which are utilized in modern strategies. The experiments demonstrate that the intrinsic locality of the CNNs and the self-attention of the Transformers induce texture-bias and shape-bias, respectively." @default.
- W4315704975 created "2023-01-12" @default.
- W4315704975 creator A5019689626 @default.
- W4315704975 creator A5035667998 @default.
- W4315704975 creator A5088907714 @default.
- W4315704975 date "2023-01-08" @default.
- W4315704975 modified "2023-09-26" @default.
- W4315704975 title "A Study on the Generality of Neural Network Structures for Monocular Depth Estimation" @default.
- W4315704975 doi "https://doi.org/10.48550/arxiv.2301.03169" @default.
- W4315704975 hasPublicationYear "2023" @default.
- W4315704975 type Work @default.
- W4315704975 citedByCount "0" @default.
- W4315704975 crossrefType "posted-content" @default.
- W4315704975 hasAuthorship W4315704975A5019689626 @default.
- W4315704975 hasAuthorship W4315704975A5035667998 @default.
- W4315704975 hasAuthorship W4315704975A5088907714 @default.
- W4315704975 hasBestOaLocation W43157049751 @default.
- W4315704975 hasConcept C119599485 @default.
- W4315704975 hasConcept C127313418 @default.
- W4315704975 hasConcept C127413603 @default.
- W4315704975 hasConcept C13280743 @default.
- W4315704975 hasConcept C134306372 @default.
- W4315704975 hasConcept C138885662 @default.
- W4315704975 hasConcept C153180895 @default.
- W4315704975 hasConcept C154945302 @default.
- W4315704975 hasConcept C15744967 @default.
- W4315704975 hasConcept C165801399 @default.
- W4315704975 hasConcept C177148314 @default.
- W4315704975 hasConcept C185798385 @default.
- W4315704975 hasConcept C2779808786 @default.
- W4315704975 hasConcept C2780767217 @default.
- W4315704975 hasConcept C31972630 @default.
- W4315704975 hasConcept C33923547 @default.
- W4315704975 hasConcept C41008148 @default.
- W4315704975 hasConcept C41895202 @default.
- W4315704975 hasConcept C50644808 @default.
- W4315704975 hasConcept C542102704 @default.
- W4315704975 hasConcept C65909025 @default.
- W4315704975 hasConcept C66322947 @default.
- W4315704975 hasConceptScore W4315704975C119599485 @default.
- W4315704975 hasConceptScore W4315704975C127313418 @default.
- W4315704975 hasConceptScore W4315704975C127413603 @default.
- W4315704975 hasConceptScore W4315704975C13280743 @default.
- W4315704975 hasConceptScore W4315704975C134306372 @default.
- W4315704975 hasConceptScore W4315704975C138885662 @default.
- W4315704975 hasConceptScore W4315704975C153180895 @default.
- W4315704975 hasConceptScore W4315704975C154945302 @default.
- W4315704975 hasConceptScore W4315704975C15744967 @default.
- W4315704975 hasConceptScore W4315704975C165801399 @default.
- W4315704975 hasConceptScore W4315704975C177148314 @default.
- W4315704975 hasConceptScore W4315704975C185798385 @default.
- W4315704975 hasConceptScore W4315704975C2779808786 @default.
- W4315704975 hasConceptScore W4315704975C2780767217 @default.
- W4315704975 hasConceptScore W4315704975C31972630 @default.
- W4315704975 hasConceptScore W4315704975C33923547 @default.
- W4315704975 hasConceptScore W4315704975C41008148 @default.
- W4315704975 hasConceptScore W4315704975C41895202 @default.
- W4315704975 hasConceptScore W4315704975C50644808 @default.
- W4315704975 hasConceptScore W4315704975C542102704 @default.
- W4315704975 hasConceptScore W4315704975C65909025 @default.
- W4315704975 hasConceptScore W4315704975C66322947 @default.
- W4315704975 hasLocation W43157049751 @default.
- W4315704975 hasOpenAccess W4315704975 @default.
- W4315704975 hasPrimaryLocation W43157049751 @default.
- W4315704975 hasRelatedWork W127013308 @default.
- W4315704975 hasRelatedWork W1628937209 @default.
- W4315704975 hasRelatedWork W1881882553 @default.
- W4315704975 hasRelatedWork W2094957557 @default.
- W4315704975 hasRelatedWork W2113039159 @default.
- W4315704975 hasRelatedWork W2971051170 @default.
- W4315704975 hasRelatedWork W3006563365 @default.
- W4315704975 hasRelatedWork W3185738386 @default.
- W4315704975 hasRelatedWork W4224011692 @default.
- W4315704975 hasRelatedWork W4225827157 @default.
- W4315704975 isParatext "false" @default.
- W4315704975 isRetracted "false" @default.
- W4315704975 workType "article" @default.