Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951628298> ?p ?o ?g. }
- W2951628298 abstract "We show how to train a Convolutional Neural Network to assign a canonical orientation to feature points given an image patch centered on the feature point. Our method improves feature point matching upon the state-of-the art and can be used in conjunction with any existing rotation sensitive descriptors. To avoid the tedious and almost impossible task of finding a target orientation to learn, we propose to use Siamese networks which implicitly find the optimal orientations during training. We also propose a new type of activation function for Neural Networks that generalizes the popular ReLU, maxout, and PReLU activation functions. This novel activation performs better for our task. We validate the effectiveness of our method extensively with four existing datasets, including two non-planar datasets, as well as our own dataset. We show that we outperform the state-of-the-art without the need of retraining for each dataset." @default.
- W2951628298 created "2019-06-27" @default.
- W2951628298 creator A5038674741 @default.
- W2951628298 creator A5049893251 @default.
- W2951628298 creator A5070382607 @default.
- W2951628298 creator A5072008141 @default.
- W2951628298 date "2015-11-13" @default.
- W2951628298 modified "2023-10-03" @default.
- W2951628298 title "Learning to Assign Orientations to Feature Points" @default.
- W2951628298 cites W1496395994 @default.
- W2951628298 cites W1606347560 @default.
- W2951628298 cites W1677182931 @default.
- W2951628298 cites W1677409904 @default.
- W2951628298 cites W1966805040 @default.
- W2951628298 cites W1995266040 @default.
- W2951628298 cites W2001790138 @default.
- W2951628298 cites W2007178811 @default.
- W2951628298 cites W2008213480 @default.
- W2951628298 cites W2028394888 @default.
- W2951628298 cites W2055685931 @default.
- W2951628298 cites W2083833836 @default.
- W2951628298 cites W2092415259 @default.
- W2951628298 cites W2095705004 @default.
- W2951628298 cites W2097934584 @default.
- W2951628298 cites W2104853049 @default.
- W2951628298 cites W2105303354 @default.
- W2951628298 cites W2119745476 @default.
- W2951628298 cites W2122015342 @default.
- W2951628298 cites W2127589108 @default.
- W2951628298 cites W2130068374 @default.
- W2951628298 cites W2139868442 @default.
- W2951628298 cites W2151103935 @default.
- W2951628298 cites W2156539399 @default.
- W2951628298 cites W2163096995 @default.
- W2951628298 cites W2169561155 @default.
- W2951628298 cites W2177274842 @default.
- W2951628298 cites W23656268 @default.
- W2951628298 cites W2616465717 @default.
- W2951628298 cites W2949213045 @default.
- W2951628298 cites W3037950864 @default.
- W2951628298 cites W2095702049 @default.
- W2951628298 hasPublicationYear "2015" @default.
- W2951628298 type Work @default.
- W2951628298 sameAs 2951628298 @default.
- W2951628298 citedByCount "2" @default.
- W2951628298 countsByYear W29516282982016 @default.
- W2951628298 countsByYear W29516282982017 @default.
- W2951628298 crossrefType "posted-content" @default.
- W2951628298 hasAuthorship W2951628298A5038674741 @default.
- W2951628298 hasAuthorship W2951628298A5049893251 @default.
- W2951628298 hasAuthorship W2951628298A5070382607 @default.
- W2951628298 hasAuthorship W2951628298A5072008141 @default.
- W2951628298 hasConcept C105795698 @default.
- W2951628298 hasConcept C119857082 @default.
- W2951628298 hasConcept C138885662 @default.
- W2951628298 hasConcept C14036430 @default.
- W2951628298 hasConcept C144133560 @default.
- W2951628298 hasConcept C153180895 @default.
- W2951628298 hasConcept C154945302 @default.
- W2951628298 hasConcept C155202549 @default.
- W2951628298 hasConcept C162324750 @default.
- W2951628298 hasConcept C16345878 @default.
- W2951628298 hasConcept C165064840 @default.
- W2951628298 hasConcept C187736073 @default.
- W2951628298 hasConcept C2524010 @default.
- W2951628298 hasConcept C2776401178 @default.
- W2951628298 hasConcept C2778712577 @default.
- W2951628298 hasConcept C2780451532 @default.
- W2951628298 hasConcept C28719098 @default.
- W2951628298 hasConcept C33923547 @default.
- W2951628298 hasConcept C38365724 @default.
- W2951628298 hasConcept C41008148 @default.
- W2951628298 hasConcept C41895202 @default.
- W2951628298 hasConcept C50644808 @default.
- W2951628298 hasConcept C74050887 @default.
- W2951628298 hasConcept C78458016 @default.
- W2951628298 hasConcept C81363708 @default.
- W2951628298 hasConcept C86803240 @default.
- W2951628298 hasConceptScore W2951628298C105795698 @default.
- W2951628298 hasConceptScore W2951628298C119857082 @default.
- W2951628298 hasConceptScore W2951628298C138885662 @default.
- W2951628298 hasConceptScore W2951628298C14036430 @default.
- W2951628298 hasConceptScore W2951628298C144133560 @default.
- W2951628298 hasConceptScore W2951628298C153180895 @default.
- W2951628298 hasConceptScore W2951628298C154945302 @default.
- W2951628298 hasConceptScore W2951628298C155202549 @default.
- W2951628298 hasConceptScore W2951628298C162324750 @default.
- W2951628298 hasConceptScore W2951628298C16345878 @default.
- W2951628298 hasConceptScore W2951628298C165064840 @default.
- W2951628298 hasConceptScore W2951628298C187736073 @default.
- W2951628298 hasConceptScore W2951628298C2524010 @default.
- W2951628298 hasConceptScore W2951628298C2776401178 @default.
- W2951628298 hasConceptScore W2951628298C2778712577 @default.
- W2951628298 hasConceptScore W2951628298C2780451532 @default.
- W2951628298 hasConceptScore W2951628298C28719098 @default.
- W2951628298 hasConceptScore W2951628298C33923547 @default.
- W2951628298 hasConceptScore W2951628298C38365724 @default.
- W2951628298 hasConceptScore W2951628298C41008148 @default.
- W2951628298 hasConceptScore W2951628298C41895202 @default.
- W2951628298 hasConceptScore W2951628298C50644808 @default.