Matches in SemOpenAlex for { <https://semopenalex.org/work/W3118942022> ?p ?o ?g. }
- W3118942022 abstract "State-of-the-art self-supervised learning approaches for monocular depth estimation usually suffer from scale ambiguity. They do not generalize well when applied on distance estimation for complex projection models such as in fisheye and omnidirectional cameras. This paper introduces a novel multi-task learning strategy to improve self-supervised monocular distance estimation on fisheye and pinhole camera images. Our contribution to this work is threefold: Firstly, we introduce a novel distance estimation network architecture using a self-attention based encoder coupled with robust semantic feature guidance to the decoder that can be trained in a one-stage fashion. Secondly, we integrate a generalized robust loss function, which improves performance significantly while removing the need for hyperparameter tuning with the reprojection loss. Finally, we reduce the artifacts caused by dynamic objects violating static world assumptions using a semantic masking strategy. We significantly improve upon the RMSE of previous work on fisheye by 25% reduction in RMSE. As there is little work on fisheye cameras, we evaluated the proposed method on KITTI using a pinhole model. We achieved state-of-the-art performance among self-supervised methods without requiring an external scale estimation." @default.
- W3118942022 created "2021-01-18" @default.
- W3118942022 creator A5002593702 @default.
- W3118942022 creator A5014764449 @default.
- W3118942022 creator A5036092986 @default.
- W3118942022 creator A5047354615 @default.
- W3118942022 creator A5063243817 @default.
- W3118942022 creator A5063282151 @default.
- W3118942022 date "2021-01-01" @default.
- W3118942022 modified "2023-10-01" @default.
- W3118942022 title "SynDistNet: Self-Supervised Monocular Fisheye Camera Distance Estimation Synergized with Semantic Segmentation for Autonomous Driving" @default.
- W3118942022 cites W1905829557 @default.
- W3118942022 cites W2133665775 @default.
- W3118942022 cites W2520707372 @default.
- W3118942022 cites W2604909019 @default.
- W3118942022 cites W2609883120 @default.
- W3118942022 cites W2889061519 @default.
- W3118942022 cites W2905536983 @default.
- W3118942022 cites W2909393256 @default.
- W3118942022 cites W2910628332 @default.
- W3118942022 cites W2934279571 @default.
- W3118942022 cites W2935854115 @default.
- W3118942022 cites W2948647700 @default.
- W3118942022 cites W2949023359 @default.
- W3118942022 cites W2952813711 @default.
- W3118942022 cites W2955863859 @default.
- W3118942022 cites W2962804601 @default.
- W3118942022 cites W2963265330 @default.
- W3118942022 cites W2963316641 @default.
- W3118942022 cites W2963412495 @default.
- W3118942022 cites W2963488291 @default.
- W3118942022 cites W2963499229 @default.
- W3118942022 cites W2963583471 @default.
- W3118942022 cites W2963652981 @default.
- W3118942022 cites W2963654727 @default.
- W3118942022 cites W2963870605 @default.
- W3118942022 cites W2963906250 @default.
- W3118942022 cites W2964089273 @default.
- W3118942022 cites W2964677288 @default.
- W3118942022 cites W2964968086 @default.
- W3118942022 cites W2967043539 @default.
- W3118942022 cites W2968529893 @default.
- W3118942022 cites W2969365860 @default.
- W3118942022 cites W2970140852 @default.
- W3118942022 cites W2981732213 @default.
- W3118942022 cites W2982014906 @default.
- W3118942022 cites W2982102242 @default.
- W3118942022 cites W2983393775 @default.
- W3118942022 cites W2985775862 @default.
- W3118942022 cites W2990874363 @default.
- W3118942022 cites W2995687155 @default.
- W3118942022 cites W2997162428 @default.
- W3118942022 cites W3004734937 @default.
- W3118942022 cites W3011843235 @default.
- W3118942022 cites W3034556105 @default.
- W3118942022 cites W3034604951 @default.
- W3118942022 cites W3034728336 @default.
- W3118942022 cites W3035004134 @default.
- W3118942022 cites W3035679448 @default.
- W3118942022 cites W3090198514 @default.
- W3118942022 doi "https://doi.org/10.1109/wacv48630.2021.00011" @default.
- W3118942022 hasPublicationYear "2021" @default.
- W3118942022 type Work @default.
- W3118942022 sameAs 3118942022 @default.
- W3118942022 citedByCount "36" @default.
- W3118942022 countsByYear W31189420222021 @default.
- W3118942022 countsByYear W31189420222022 @default.
- W3118942022 countsByYear W31189420222023 @default.
- W3118942022 crossrefType "proceedings-article" @default.
- W3118942022 hasAuthorship W3118942022A5002593702 @default.
- W3118942022 hasAuthorship W3118942022A5014764449 @default.
- W3118942022 hasAuthorship W3118942022A5036092986 @default.
- W3118942022 hasAuthorship W3118942022A5047354615 @default.
- W3118942022 hasAuthorship W3118942022A5063243817 @default.
- W3118942022 hasAuthorship W3118942022A5063282151 @default.
- W3118942022 hasBestOaLocation W31189420222 @default.
- W3118942022 hasConcept C104317684 @default.
- W3118942022 hasConcept C105795698 @default.
- W3118942022 hasConcept C139945424 @default.
- W3118942022 hasConcept C153180895 @default.
- W3118942022 hasConcept C154945302 @default.
- W3118942022 hasConcept C158829959 @default.
- W3118942022 hasConcept C185592680 @default.
- W3118942022 hasConcept C31972630 @default.
- W3118942022 hasConcept C33923547 @default.
- W3118942022 hasConcept C41008148 @default.
- W3118942022 hasConcept C55493867 @default.
- W3118942022 hasConcept C63479239 @default.
- W3118942022 hasConcept C65909025 @default.
- W3118942022 hasConcept C89600930 @default.
- W3118942022 hasConceptScore W3118942022C104317684 @default.
- W3118942022 hasConceptScore W3118942022C105795698 @default.
- W3118942022 hasConceptScore W3118942022C139945424 @default.
- W3118942022 hasConceptScore W3118942022C153180895 @default.
- W3118942022 hasConceptScore W3118942022C154945302 @default.
- W3118942022 hasConceptScore W3118942022C158829959 @default.
- W3118942022 hasConceptScore W3118942022C185592680 @default.
- W3118942022 hasConceptScore W3118942022C31972630 @default.
- W3118942022 hasConceptScore W3118942022C33923547 @default.
- W3118942022 hasConceptScore W3118942022C41008148 @default.