Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384161763> ?p ?o ?g. }
- W4384161763 endingPage "78261" @default.
- W4384161763 startingPage "78251" @default.
- W4384161763 abstract "Image-guided depth completion aims to generate dense depth maps from sparse depth maps guided by their corresponding color (RGB) images. In this paper, we propose deep sparse depth completion using multi-affinity matrix. Recently, spatial propagation networks (SPNs) are used to refine depth maps obtained by initial depth completion. However, they use the same affinity matrix even in multiple iterations that has a limit to improving performance, which is not effective in considering the relationship between adjacent pixels. Thus, we replace it with a multi-affinity matrix to represent the relationship between an output pixel and its neighboring ones. Moreover, deep neural networks can effectively fuse features from two different modalities based on confidence maps. Inspired by dynamic guided filtering, we use a convolutional spatial propagation network (CSPN) to fuse multi-modal features at multiple stages. When training the color branch of the proposed network, we adopt supervised learning that constrains the output features of all layers in the decoder. Since spatially varying features are required for multi-modal feature fusion, the proposed network produces adaptive affinity features using a single decoder. Experimental results on KITTI and NYU-v2 datasets show that the multi-affinity matrix represents the dependency between pixels and predicts accurate depth values by calculating the weights between neighboring pixels. The proposed network based on multi-affinity matrix achieves state-of-the-art performance in terms of root mean square error (RMSE) and mean absolute error (MAE)." @default.
- W4384161763 created "2023-07-14" @default.
- W4384161763 creator A5003623628 @default.
- W4384161763 creator A5060189619 @default.
- W4384161763 creator A5066877556 @default.
- W4384161763 date "2023-01-01" @default.
- W4384161763 modified "2023-09-26" @default.
- W4384161763 title "Deep Sparse Depth Completion Using Multi-Affinity Matrix" @default.
- W4384161763 cites W2005441409 @default.
- W4384161763 cites W2062902010 @default.
- W4384161763 cites W2117248802 @default.
- W4384161763 cites W2153388956 @default.
- W4384161763 cites W2178922201 @default.
- W4384161763 cites W2194775991 @default.
- W4384161763 cites W2531409750 @default.
- W4384161763 cites W2807828983 @default.
- W4384161763 cites W2885093229 @default.
- W4384161763 cites W2886851716 @default.
- W4384161763 cites W2899479761 @default.
- W4384161763 cites W2963045776 @default.
- W4384161763 cites W2963316641 @default.
- W4384161763 cites W2963416674 @default.
- W4384161763 cites W2963867516 @default.
- W4384161763 cites W2964326562 @default.
- W4384161763 cites W2998031326 @default.
- W4384161763 cites W2998293366 @default.
- W4384161763 cites W3034421924 @default.
- W4384161763 cites W3091564232 @default.
- W4384161763 cites W3109128945 @default.
- W4384161763 cites W3110653837 @default.
- W4384161763 cites W3165495321 @default.
- W4384161763 cites W3175561084 @default.
- W4384161763 cites W3206335707 @default.
- W4384161763 cites W4312566889 @default.
- W4384161763 doi "https://doi.org/10.1109/access.2023.3295133" @default.
- W4384161763 hasPublicationYear "2023" @default.
- W4384161763 type Work @default.
- W4384161763 citedByCount "0" @default.
- W4384161763 crossrefType "journal-article" @default.
- W4384161763 hasAuthorship W4384161763A5003623628 @default.
- W4384161763 hasAuthorship W4384161763A5060189619 @default.
- W4384161763 hasAuthorship W4384161763A5066877556 @default.
- W4384161763 hasBestOaLocation W43841617631 @default.
- W4384161763 hasConcept C105795698 @default.
- W4384161763 hasConcept C106487976 @default.
- W4384161763 hasConcept C108583219 @default.
- W4384161763 hasConcept C11413529 @default.
- W4384161763 hasConcept C119599485 @default.
- W4384161763 hasConcept C121332964 @default.
- W4384161763 hasConcept C127413603 @default.
- W4384161763 hasConcept C138885662 @default.
- W4384161763 hasConcept C139945424 @default.
- W4384161763 hasConcept C141353440 @default.
- W4384161763 hasConcept C153180895 @default.
- W4384161763 hasConcept C154945302 @default.
- W4384161763 hasConcept C159985019 @default.
- W4384161763 hasConcept C160633673 @default.
- W4384161763 hasConcept C163716315 @default.
- W4384161763 hasConcept C192562407 @default.
- W4384161763 hasConcept C2776401178 @default.
- W4384161763 hasConcept C33923547 @default.
- W4384161763 hasConcept C41008148 @default.
- W4384161763 hasConcept C41895202 @default.
- W4384161763 hasConcept C56372850 @default.
- W4384161763 hasConcept C62520636 @default.
- W4384161763 hasConcept C82990744 @default.
- W4384161763 hasConceptScore W4384161763C105795698 @default.
- W4384161763 hasConceptScore W4384161763C106487976 @default.
- W4384161763 hasConceptScore W4384161763C108583219 @default.
- W4384161763 hasConceptScore W4384161763C11413529 @default.
- W4384161763 hasConceptScore W4384161763C119599485 @default.
- W4384161763 hasConceptScore W4384161763C121332964 @default.
- W4384161763 hasConceptScore W4384161763C127413603 @default.
- W4384161763 hasConceptScore W4384161763C138885662 @default.
- W4384161763 hasConceptScore W4384161763C139945424 @default.
- W4384161763 hasConceptScore W4384161763C141353440 @default.
- W4384161763 hasConceptScore W4384161763C153180895 @default.
- W4384161763 hasConceptScore W4384161763C154945302 @default.
- W4384161763 hasConceptScore W4384161763C159985019 @default.
- W4384161763 hasConceptScore W4384161763C160633673 @default.
- W4384161763 hasConceptScore W4384161763C163716315 @default.
- W4384161763 hasConceptScore W4384161763C192562407 @default.
- W4384161763 hasConceptScore W4384161763C2776401178 @default.
- W4384161763 hasConceptScore W4384161763C33923547 @default.
- W4384161763 hasConceptScore W4384161763C41008148 @default.
- W4384161763 hasConceptScore W4384161763C41895202 @default.
- W4384161763 hasConceptScore W4384161763C56372850 @default.
- W4384161763 hasConceptScore W4384161763C62520636 @default.
- W4384161763 hasConceptScore W4384161763C82990744 @default.
- W4384161763 hasFunder F4320321001 @default.
- W4384161763 hasLocation W43841617631 @default.
- W4384161763 hasOpenAccess W4384161763 @default.
- W4384161763 hasPrimaryLocation W43841617631 @default.
- W4384161763 hasRelatedWork W2090093270 @default.
- W4384161763 hasRelatedWork W2136485282 @default.
- W4384161763 hasRelatedWork W2290563756 @default.
- W4384161763 hasRelatedWork W2546871836 @default.
- W4384161763 hasRelatedWork W2769066916 @default.