Matches in SemOpenAlex for { <https://semopenalex.org/work/W3049060039> ?p ?o ?g. }
- W3049060039 abstract "Self-supervised learning has emerged as a powerful tool for depth and ego-motion estimation, leading to state-of-the-art results on benchmark datasets. However, one significant limitation shared by current methods is the assumption of a known parametric camera model -- usually the standard pinhole geometry -- leading to failure when applied to imaging systems that deviate significantly from this assumption (e.g., catadioptric cameras or underwater imaging). In this work, we show that self-supervision can be used to learn accurate depth and ego-motion estimation without prior knowledge of the camera model. Inspired by the geometric model of Grossberg and Nayar, we introduce Neural Ray Surfaces (NRS), convolutional networks that represent pixel-wise projection rays, approximating a wide range of cameras. NRS are fully differentiable and can be learned end-to-end from unlabeled raw videos. We demonstrate the use of NRS for self-supervised learning of visual odometry and depth estimation from raw videos obtained using a wide variety of camera systems, including pinhole, fisheye, and catadioptric." @default.
- W3049060039 created "2020-08-21" @default.
- W3049060039 creator A5009207422 @default.
- W3049060039 creator A5037522398 @default.
- W3049060039 creator A5042694706 @default.
- W3049060039 creator A5063755961 @default.
- W3049060039 creator A5064302017 @default.
- W3049060039 creator A5075018873 @default.
- W3049060039 creator A5084499878 @default.
- W3049060039 date "2020-08-15" @default.
- W3049060039 modified "2023-10-18" @default.
- W3049060039 title "Neural Ray Surfaces for Self-Supervised Learning of Depth and Ego-motion" @default.
- W3049060039 cites W1227273861 @default.
- W3049060039 cites W125693051 @default.
- W3049060039 cites W129896452 @default.
- W3049060039 cites W1899309388 @default.
- W3049060039 cites W1989310712 @default.
- W3049060039 cites W2033819227 @default.
- W3049060039 cites W2095705004 @default.
- W3049060039 cites W2097823282 @default.
- W3049060039 cites W2098500213 @default.
- W3049060039 cites W2115579991 @default.
- W3049060039 cites W2118341165 @default.
- W3049060039 cites W2133665775 @default.
- W3049060039 cites W2169887425 @default.
- W3049060039 cites W2411681238 @default.
- W3049060039 cites W2436453945 @default.
- W3049060039 cites W2471239744 @default.
- W3049060039 cites W2520707372 @default.
- W3049060039 cites W2608018946 @default.
- W3049060039 cites W2609883120 @default.
- W3049060039 cites W2764315368 @default.
- W3049060039 cites W2767981409 @default.
- W3049060039 cites W2776251693 @default.
- W3049060039 cites W2777162184 @default.
- W3049060039 cites W2897941598 @default.
- W3049060039 cites W2899771611 @default.
- W3049060039 cites W2903872112 @default.
- W3049060039 cites W2918922367 @default.
- W3049060039 cites W2949117887 @default.
- W3049060039 cites W2954174912 @default.
- W3049060039 cites W2962816904 @default.
- W3049060039 cites W2963285578 @default.
- W3049060039 cites W2963583471 @default.
- W3049060039 cites W2963906250 @default.
- W3049060039 cites W2964121744 @default.
- W3049060039 cites W2979318013 @default.
- W3049060039 cites W2981732213 @default.
- W3049060039 cites W2982102242 @default.
- W3049060039 cites W2985032069 @default.
- W3049060039 cites W2991215750 @default.
- W3049060039 cites W2995884594 @default.
- W3049060039 cites W3014263713 @default.
- W3049060039 cites W3034321406 @default.
- W3049060039 cites W3034604951 @default.
- W3049060039 cites W3035575130 @default.
- W3049060039 hasPublicationYear "2020" @default.
- W3049060039 type Work @default.
- W3049060039 sameAs 3049060039 @default.
- W3049060039 citedByCount "0" @default.
- W3049060039 crossrefType "posted-content" @default.
- W3049060039 hasAuthorship W3049060039A5009207422 @default.
- W3049060039 hasAuthorship W3049060039A5037522398 @default.
- W3049060039 hasAuthorship W3049060039A5042694706 @default.
- W3049060039 hasAuthorship W3049060039A5063755961 @default.
- W3049060039 hasAuthorship W3049060039A5064302017 @default.
- W3049060039 hasAuthorship W3049060039A5075018873 @default.
- W3049060039 hasAuthorship W3049060039A5084499878 @default.
- W3049060039 hasConcept C10161872 @default.
- W3049060039 hasConcept C11413529 @default.
- W3049060039 hasConcept C127413603 @default.
- W3049060039 hasConcept C13280743 @default.
- W3049060039 hasConcept C142362112 @default.
- W3049060039 hasConcept C153349607 @default.
- W3049060039 hasConcept C15336307 @default.
- W3049060039 hasConcept C153396827 @default.
- W3049060039 hasConcept C154945302 @default.
- W3049060039 hasConcept C160633673 @default.
- W3049060039 hasConcept C185798385 @default.
- W3049060039 hasConcept C205649164 @default.
- W3049060039 hasConcept C2776374813 @default.
- W3049060039 hasConcept C31972630 @default.
- W3049060039 hasConcept C41008148 @default.
- W3049060039 hasConcept C50644808 @default.
- W3049060039 hasConcept C57493831 @default.
- W3049060039 hasConcept C65909025 @default.
- W3049060039 hasConcept C78762247 @default.
- W3049060039 hasConcept C81363708 @default.
- W3049060039 hasConceptScore W3049060039C10161872 @default.
- W3049060039 hasConceptScore W3049060039C11413529 @default.
- W3049060039 hasConceptScore W3049060039C127413603 @default.
- W3049060039 hasConceptScore W3049060039C13280743 @default.
- W3049060039 hasConceptScore W3049060039C142362112 @default.
- W3049060039 hasConceptScore W3049060039C153349607 @default.
- W3049060039 hasConceptScore W3049060039C15336307 @default.
- W3049060039 hasConceptScore W3049060039C153396827 @default.
- W3049060039 hasConceptScore W3049060039C154945302 @default.
- W3049060039 hasConceptScore W3049060039C160633673 @default.
- W3049060039 hasConceptScore W3049060039C185798385 @default.
- W3049060039 hasConceptScore W3049060039C205649164 @default.