Matches in SemOpenAlex for { <https://semopenalex.org/work/W3207920912> ?p ?o ?g. }
- W3207920912 abstract "In the field of visual ego-motion estimation for Micro Air Vehicles (MAVs), fast maneuvers stay challenging mainly because of the big visual disparity and motion blur. In the pursuit of higher robustness, we study convolutional neural networks (CNNs) that predict the relative pose between subsequent images from a fast-moving monocular camera facing a planar scene. Aided by the Inertial Measurement Unit (IMU), we mainly focus on translational motion. The networks we study have similar small model sizes (around 1.35MB) and high inference speeds (around 10 milliseconds on a mobile GPU). Images for training and testing have realistic motion blur. Departing from a network framework that iteratively warps the first image to match the second with cascaded network blocks, we study different network architectures and training strategies. Simulated datasets and a self-collected MAV flight dataset are used for evaluation. The proposed setup shows better accuracy over existing networks and traditional feature-point-based methods during fast maneuvers. Moreover, self-supervised learning outperforms supervised learning. Videos and open-sourced code are available at https://github. com/tudelft/PoseNet_Planar" @default.
- W3207920912 created "2021-10-25" @default.
- W3207920912 creator A5015108690 @default.
- W3207920912 creator A5032638027 @default.
- W3207920912 date "2021-05-30" @default.
- W3207920912 modified "2023-09-25" @default.
- W3207920912 title "CNN-based Ego-Motion Estimation for Fast MAV Maneuve" @default.
- W3207920912 cites W1533861849 @default.
- W3207920912 cites W1677409904 @default.
- W3207920912 cites W1861492603 @default.
- W3207920912 cites W1970504153 @default.
- W3207920912 cites W2024763827 @default.
- W3207920912 cites W2069295011 @default.
- W3207920912 cites W2117228865 @default.
- W3207920912 cites W2130103520 @default.
- W3207920912 cites W2150066425 @default.
- W3207920912 cites W2155302366 @default.
- W3207920912 cites W2160146756 @default.
- W3207920912 cites W2164198722 @default.
- W3207920912 cites W2200124539 @default.
- W3207920912 cites W2216550548 @default.
- W3207920912 cites W2220063164 @default.
- W3207920912 cites W2302255633 @default.
- W3207920912 cites W2396274919 @default.
- W3207920912 cites W2562066862 @default.
- W3207920912 cites W2592936284 @default.
- W3207920912 cites W2609883120 @default.
- W3207920912 cites W2745859992 @default.
- W3207920912 cites W2765767940 @default.
- W3207920912 cites W2910881200 @default.
- W3207920912 cites W2962987986 @default.
- W3207920912 cites W2963446712 @default.
- W3207920912 cites W2963782415 @default.
- W3207920912 cites W2964121744 @default.
- W3207920912 cites W2964314455 @default.
- W3207920912 cites W2964968086 @default.
- W3207920912 cites W2970971581 @default.
- W3207920912 cites W2971000934 @default.
- W3207920912 cites W2981732213 @default.
- W3207920912 cites W3000760939 @default.
- W3207920912 cites W3035025630 @default.
- W3207920912 cites W3040334549 @default.
- W3207920912 cites W3103648783 @default.
- W3207920912 cites W603908379 @default.
- W3207920912 doi "https://doi.org/10.1109/icra48506.2021.9561714" @default.
- W3207920912 hasPublicationYear "2021" @default.
- W3207920912 type Work @default.
- W3207920912 sameAs 3207920912 @default.
- W3207920912 citedByCount "0" @default.
- W3207920912 crossrefType "proceedings-article" @default.
- W3207920912 hasAuthorship W3207920912A5015108690 @default.
- W3207920912 hasAuthorship W3207920912A5032638027 @default.
- W3207920912 hasBestOaLocation W32079209122 @default.
- W3207920912 hasConcept C10161872 @default.
- W3207920912 hasConcept C104317684 @default.
- W3207920912 hasConcept C108583219 @default.
- W3207920912 hasConcept C115961682 @default.
- W3207920912 hasConcept C120665830 @default.
- W3207920912 hasConcept C121332964 @default.
- W3207920912 hasConcept C154945302 @default.
- W3207920912 hasConcept C185592680 @default.
- W3207920912 hasConcept C192209626 @default.
- W3207920912 hasConcept C2776214188 @default.
- W3207920912 hasConcept C2777708103 @default.
- W3207920912 hasConcept C31972630 @default.
- W3207920912 hasConcept C41008148 @default.
- W3207920912 hasConcept C55493867 @default.
- W3207920912 hasConcept C63479239 @default.
- W3207920912 hasConcept C65909025 @default.
- W3207920912 hasConcept C79061980 @default.
- W3207920912 hasConcept C81363708 @default.
- W3207920912 hasConceptScore W3207920912C10161872 @default.
- W3207920912 hasConceptScore W3207920912C104317684 @default.
- W3207920912 hasConceptScore W3207920912C108583219 @default.
- W3207920912 hasConceptScore W3207920912C115961682 @default.
- W3207920912 hasConceptScore W3207920912C120665830 @default.
- W3207920912 hasConceptScore W3207920912C121332964 @default.
- W3207920912 hasConceptScore W3207920912C154945302 @default.
- W3207920912 hasConceptScore W3207920912C185592680 @default.
- W3207920912 hasConceptScore W3207920912C192209626 @default.
- W3207920912 hasConceptScore W3207920912C2776214188 @default.
- W3207920912 hasConceptScore W3207920912C2777708103 @default.
- W3207920912 hasConceptScore W3207920912C31972630 @default.
- W3207920912 hasConceptScore W3207920912C41008148 @default.
- W3207920912 hasConceptScore W3207920912C55493867 @default.
- W3207920912 hasConceptScore W3207920912C63479239 @default.
- W3207920912 hasConceptScore W3207920912C65909025 @default.
- W3207920912 hasConceptScore W3207920912C79061980 @default.
- W3207920912 hasConceptScore W3207920912C81363708 @default.
- W3207920912 hasLocation W32079209121 @default.
- W3207920912 hasLocation W32079209122 @default.
- W3207920912 hasOpenAccess W3207920912 @default.
- W3207920912 hasPrimaryLocation W32079209121 @default.
- W3207920912 hasRelatedWork W1970628095 @default.
- W3207920912 hasRelatedWork W2009198678 @default.
- W3207920912 hasRelatedWork W2325200052 @default.
- W3207920912 hasRelatedWork W2562524049 @default.
- W3207920912 hasRelatedWork W2810697792 @default.
- W3207920912 hasRelatedWork W2980357211 @default.
- W3207920912 hasRelatedWork W3118809760 @default.