FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2904382404> ?p ?o ?g. }

• W2904382404 abstract "In this work, we propose a method that combines unsupervised deep learning predictions for optical flow and monocular disparity with a model based optimization procedure for instantaneous camera pose. Given the flow and disparity predictions from the network, we apply a RANSAC outlier rejection scheme to find an inlier set of flows and disparities, which we use to solve for the relative camera pose in a least squares fashion. We show that this pipeline is fully differentiable, allowing us to combine the pose with the network outputs as an additional unsupervised training loss to further refine the predicted flows and disparities. This method not only allows us to directly regress relative pose from the network outputs, but also automatically segments away pixels that do not fit the rigid scene assumptions that many unsupervised structure from motion methods apply, such as on independently moving objects. We evaluate our method on the KITTI dataset, and demonstrate state of the art results, even in the presence of challenging independently moving objects." @default.
• W2904382404 created "2018-12-22" @default.
• W2904382404 creator A5000779441 @default.
• W2904382404 creator A5018724340 @default.
• W2904382404 creator A5050660826 @default.
• W2904382404 creator A5059471389 @default.
• W2904382404 creator A5070302005 @default.
• W2904382404 date "2018-12-20" @default.
• W2904382404 modified "2023-09-23" @default.
• W2904382404 title "Robustness Meets Deep Learning: An End-to-End Hybrid Pipeline for Unsupervised Learning of Egomotion" @default.
• W2904382404 cites W2085261163 @default.
• W2904382404 cites W2115579991 @default.
• W2904382404 cites W2133665775 @default.
• W2904382404 cites W2150066425 @default.
• W2904382404 cites W2171740948 @default.
• W2904382404 cites W2194775991 @default.
• W2904382404 cites W2286655030 @default.
• W2904382404 cites W2300779272 @default.
• W2904382404 cites W2474281075 @default.
• W2904382404 cites W2507953016 @default.
• W2904382404 cites W2520707372 @default.
• W2904382404 cites W2556455135 @default.
• W2904382404 cites W2608018946 @default.
• W2904382404 cites W2609883120 @default.
• W2904382404 cites W2751625733 @default.
• W2904382404 cites W2770902190 @default.
• W2904382404 cites W2804788208 @default.
• W2904382404 cites W2830339951 @default.
• W2904382404 cites W2883702102 @default.
• W2904382404 cites W2949634581 @default.
• W2904382404 cites W2950471768 @default.
• W2904382404 cites W2951309005 @default.
• W2904382404 cites W2962816904 @default.
• W2904382404 cites W2963265330 @default.
• W2904382404 cites W2963583471 @default.
• W2904382404 cites W2963654727 @default.
• W2904382404 cites W2963782415 @default.
• W2904382404 cites W2963891416 @default.
• W2904382404 cites W2964314455 @default.
• W2904382404 cites W2964968086 @default.
• W2904382404 cites W603908379 @default.
• W2904382404 cites W764651262 @default.
• W2904382404 hasPublicationYear "2018" @default.
• W2904382404 type Work @default.
• W2904382404 sameAs 2904382404 @default.
• W2904382404 citedByCount "6" @default.
• W2904382404 countsByYear W29043824042019 @default.
• W2904382404 countsByYear W29043824042020 @default.
• W2904382404 countsByYear W29043824042021 @default.
• W2904382404 crossrefType "posted-content" @default.
• W2904382404 hasAuthorship W2904382404A5000779441 @default.
• W2904382404 hasAuthorship W2904382404A5018724340 @default.
• W2904382404 hasAuthorship W2904382404A5050660826 @default.
• W2904382404 hasAuthorship W2904382404A5059471389 @default.
• W2904382404 hasAuthorship W2904382404A5070302005 @default.
• W2904382404 hasConcept C104317684 @default.
• W2904382404 hasConcept C108583219 @default.
• W2904382404 hasConcept C114744707 @default.
• W2904382404 hasConcept C115961682 @default.
• W2904382404 hasConcept C153180895 @default.
• W2904382404 hasConcept C154945302 @default.
• W2904382404 hasConcept C155542232 @default.
• W2904382404 hasConcept C160633673 @default.
• W2904382404 hasConcept C185592680 @default.
• W2904382404 hasConcept C31972630 @default.
• W2904382404 hasConcept C41008148 @default.
• W2904382404 hasConcept C55493867 @default.
• W2904382404 hasConcept C63479239 @default.
• W2904382404 hasConcept C65909025 @default.
• W2904382404 hasConcept C79337645 @default.
• W2904382404 hasConcept C8038995 @default.
• W2904382404 hasConceptScore W2904382404C104317684 @default.
• W2904382404 hasConceptScore W2904382404C108583219 @default.
• W2904382404 hasConceptScore W2904382404C114744707 @default.
• W2904382404 hasConceptScore W2904382404C115961682 @default.
• W2904382404 hasConceptScore W2904382404C153180895 @default.
• W2904382404 hasConceptScore W2904382404C154945302 @default.
• W2904382404 hasConceptScore W2904382404C155542232 @default.
• W2904382404 hasConceptScore W2904382404C160633673 @default.
• W2904382404 hasConceptScore W2904382404C185592680 @default.
• W2904382404 hasConceptScore W2904382404C31972630 @default.
• W2904382404 hasConceptScore W2904382404C41008148 @default.
• W2904382404 hasConceptScore W2904382404C55493867 @default.
• W2904382404 hasConceptScore W2904382404C63479239 @default.
• W2904382404 hasConceptScore W2904382404C65909025 @default.
• W2904382404 hasConceptScore W2904382404C79337645 @default.
• W2904382404 hasConceptScore W2904382404C8038995 @default.
• W2904382404 hasLocation W29043824041 @default.
• W2904382404 hasOpenAccess W2904382404 @default.
• W2904382404 hasPrimaryLocation W29043824041 @default.
• W2904382404 hasRelatedWork W2115579991 @default.
• W2904382404 hasRelatedWork W2133665775 @default.
• W2904382404 hasRelatedWork W2150066425 @default.
• W2904382404 hasRelatedWork W2168676389 @default.
• W2904382404 hasRelatedWork W2171740948 @default.
• W2904382404 hasRelatedWork W2474281075 @default.
• W2904382404 hasRelatedWork W2520707372 @default.
• W2904382404 hasRelatedWork W2609883120 @default.
• W2904382404 hasRelatedWork W2830339951 @default.
• W2904382404 hasRelatedWork W2892197942 @default.

• next