Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385800744> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W4385800744 abstract "The ubiquitous multi-camera setup on modern autonomous vehicles provides an opportunity to construct surround-view depth. Existing methods, however, either perform independent monocular depth estimations on each camera or rely on computationally heavy self attention mechanisms. In this paper, we propose a novel guided attention architecture, EGA-Depth, which can improve both the efficiency and accuracy of self-supervised multi-camera depth estimation. More specifically, for each camera, we use its perspective view as the query to cross-reference its neighboring views to derive informative features for this camera view. This allows the model to perform attention only across views with considerable overlaps and avoid the costly computations of standard self-attention. Given its efficiency, EGA-Depth enables us to exploit higher-resolution visual features, leading to improved accuracy. Furthermore, EGA-Depth can incorporate more frames from previous time steps as it scales linearly w.r.t. the number of views and frames. Extensive experiments on two challenging autonomous driving benchmarks nuScenes and DDAD demonstrate the efficacy of our proposed EGA-Depth and show that it achieves the new state-of-the-art in self-supervised multi-camera depth estimation." @default.
- W4385800744 created "2023-08-15" @default.
- W4385800744 creator A5015866633 @default.
- W4385800744 creator A5016524358 @default.
- W4385800744 creator A5038550641 @default.
- W4385800744 creator A5053652981 @default.
- W4385800744 date "2023-06-01" @default.
- W4385800744 modified "2023-10-16" @default.
- W4385800744 title "EGA-Depth: Efficient Guided Attention for Self-Supervised Multi-Camera Depth Estimation" @default.
- W4385800744 cites W2008706659 @default.
- W4385800744 cites W2108134361 @default.
- W4385800744 cites W2108598243 @default.
- W4385800744 cites W2133665775 @default.
- W4385800744 cites W2194775991 @default.
- W4385800744 cites W2609883120 @default.
- W4385800744 cites W2963412495 @default.
- W4385800744 cites W2963488291 @default.
- W4385800744 cites W2982102242 @default.
- W4385800744 cites W2985775862 @default.
- W4385800744 cites W3014263713 @default.
- W4385800744 cites W3034403088 @default.
- W4385800744 cites W3034604951 @default.
- W4385800744 cites W3035574168 @default.
- W4385800744 cites W3060975791 @default.
- W4385800744 cites W3107389224 @default.
- W4385800744 cites W3138516171 @default.
- W4385800744 cites W3141975725 @default.
- W4385800744 cites W3173365702 @default.
- W4385800744 cites W3175682855 @default.
- W4385800744 cites W4206706211 @default.
- W4385800744 cites W4210254627 @default.
- W4385800744 cites W4214614183 @default.
- W4385800744 cites W4312572946 @default.
- W4385800744 cites W4312769131 @default.
- W4385800744 cites W4312819733 @default.
- W4385800744 doi "https://doi.org/10.1109/cvprw59228.2023.00017" @default.
- W4385800744 hasPublicationYear "2023" @default.
- W4385800744 type Work @default.
- W4385800744 citedByCount "0" @default.
- W4385800744 crossrefType "proceedings-article" @default.
- W4385800744 hasAuthorship W4385800744A5015866633 @default.
- W4385800744 hasAuthorship W4385800744A5016524358 @default.
- W4385800744 hasAuthorship W4385800744A5038550641 @default.
- W4385800744 hasAuthorship W4385800744A5053652981 @default.
- W4385800744 hasBestOaLocation W43858007442 @default.
- W4385800744 hasConcept C110898773 @default.
- W4385800744 hasConcept C11413529 @default.
- W4385800744 hasConcept C12713177 @default.
- W4385800744 hasConcept C154945302 @default.
- W4385800744 hasConcept C165696696 @default.
- W4385800744 hasConcept C199360897 @default.
- W4385800744 hasConcept C2780801425 @default.
- W4385800744 hasConcept C31972630 @default.
- W4385800744 hasConcept C38652104 @default.
- W4385800744 hasConcept C41008148 @default.
- W4385800744 hasConcept C45374587 @default.
- W4385800744 hasConcept C65909025 @default.
- W4385800744 hasConcept C94816000 @default.
- W4385800744 hasConceptScore W4385800744C110898773 @default.
- W4385800744 hasConceptScore W4385800744C11413529 @default.
- W4385800744 hasConceptScore W4385800744C12713177 @default.
- W4385800744 hasConceptScore W4385800744C154945302 @default.
- W4385800744 hasConceptScore W4385800744C165696696 @default.
- W4385800744 hasConceptScore W4385800744C199360897 @default.
- W4385800744 hasConceptScore W4385800744C2780801425 @default.
- W4385800744 hasConceptScore W4385800744C31972630 @default.
- W4385800744 hasConceptScore W4385800744C38652104 @default.
- W4385800744 hasConceptScore W4385800744C41008148 @default.
- W4385800744 hasConceptScore W4385800744C45374587 @default.
- W4385800744 hasConceptScore W4385800744C65909025 @default.
- W4385800744 hasConceptScore W4385800744C94816000 @default.
- W4385800744 hasLocation W43858007441 @default.
- W4385800744 hasLocation W43858007442 @default.
- W4385800744 hasOpenAccess W4385800744 @default.
- W4385800744 hasPrimaryLocation W43858007441 @default.
- W4385800744 hasRelatedWork W2010809152 @default.
- W4385800744 hasRelatedWork W2027392171 @default.
- W4385800744 hasRelatedWork W2055841316 @default.
- W4385800744 hasRelatedWork W2092917442 @default.
- W4385800744 hasRelatedWork W2120510242 @default.
- W4385800744 hasRelatedWork W2386208773 @default.
- W4385800744 hasRelatedWork W2793501283 @default.
- W4385800744 hasRelatedWork W2964433127 @default.
- W4385800744 hasRelatedWork W3080417936 @default.
- W4385800744 hasRelatedWork W4287684818 @default.
- W4385800744 isParatext "false" @default.
- W4385800744 isRetracted "false" @default.
- W4385800744 workType "article" @default.