Matches in SemOpenAlex for { <https://semopenalex.org/work/W3046455422> ?p ?o ?g. }
- W3046455422 abstract "In this paper, we propose a method to map data from a Grassmann manifold to a vector space while maximizing discrimination capability for classification. Subspaces are a practical and robust representation for image set recognition. However, as they exist on a Grassmann manifold, machine learning tools constructed on Euclidean geometry cannot be promptly utilized. Recently, methods to construct end-to-end learnable models for subspaces are starting to be explored, but they require multiple matrix decompositions and can be hard to compute and extend. Therefore we introduce a layer to map Grassmann manifold-valued data to vector space, in such a way that it can be seamlessly used as a layer along with other powerful tools defined on Euclidean space. The key idea of our method is to formulate the manifold logarithmic map (log) as a learnable model, where we seek to learn a tangency point that minimizes a loss function with respect to the data. The log effectively transforms a manifold point into a tangent vector. This log model can be learned with Riemannian stochastic gradient descent on the target manifold. We demonstrate the effectiveness of our proposed method on the applications of hand shape recognition, face identification and facial emotion recognition." @default.
- W3046455422 created "2020-08-07" @default.
- W3046455422 creator A5004409803 @default.
- W3046455422 creator A5024888646 @default.
- W3046455422 creator A5064373853 @default.
- W3046455422 creator A5069526951 @default.
- W3046455422 creator A5073772009 @default.
- W3046455422 date "2020-06-01" @default.
- W3046455422 modified "2023-09-25" @default.
- W3046455422 title "An Interface between Grassmann manifolds and vector spaces" @default.
- W3046455422 cites W1551774182 @default.
- W3046455422 cites W1606858007 @default.
- W3046455422 cites W1862697533 @default.
- W3046455422 cites W1900408618 @default.
- W3046455422 cites W1922045146 @default.
- W3046455422 cites W1985188514 @default.
- W3046455422 cites W1992227055 @default.
- W3046455422 cites W1996939238 @default.
- W3046455422 cites W2084146405 @default.
- W3046455422 cites W2103924867 @default.
- W3046455422 cites W2123582174 @default.
- W3046455422 cites W2126017757 @default.
- W3046455422 cites W2134860945 @default.
- W3046455422 cites W2147238549 @default.
- W3046455422 cites W2147631863 @default.
- W3046455422 cites W2149338181 @default.
- W3046455422 cites W2150600350 @default.
- W3046455422 cites W2160888808 @default.
- W3046455422 cites W2194775991 @default.
- W3046455422 cites W2432877789 @default.
- W3046455422 cites W2486791310 @default.
- W3046455422 cites W2517500902 @default.
- W3046455422 cites W2551687757 @default.
- W3046455422 cites W2574111200 @default.
- W3046455422 cites W2737267232 @default.
- W3046455422 cites W2770659060 @default.
- W3046455422 cites W2895643041 @default.
- W3046455422 cites W2903028847 @default.
- W3046455422 cites W2917342980 @default.
- W3046455422 cites W3097096317 @default.
- W3046455422 cites W3102317997 @default.
- W3046455422 doi "https://doi.org/10.1109/cvprw50498.2020.00431" @default.
- W3046455422 hasPublicationYear "2020" @default.
- W3046455422 type Work @default.
- W3046455422 sameAs 3046455422 @default.
- W3046455422 citedByCount "4" @default.
- W3046455422 countsByYear W30464554222021 @default.
- W3046455422 countsByYear W30464554222022 @default.
- W3046455422 crossrefType "proceedings-article" @default.
- W3046455422 hasAuthorship W3046455422A5004409803 @default.
- W3046455422 hasAuthorship W3046455422A5024888646 @default.
- W3046455422 hasAuthorship W3046455422A5064373853 @default.
- W3046455422 hasAuthorship W3046455422A5069526951 @default.
- W3046455422 hasAuthorship W3046455422A5073772009 @default.
- W3046455422 hasConcept C11413529 @default.
- W3046455422 hasConcept C12362212 @default.
- W3046455422 hasConcept C127413603 @default.
- W3046455422 hasConcept C138187205 @default.
- W3046455422 hasConcept C151876577 @default.
- W3046455422 hasConcept C153120616 @default.
- W3046455422 hasConcept C153180895 @default.
- W3046455422 hasConcept C154945302 @default.
- W3046455422 hasConcept C157157409 @default.
- W3046455422 hasConcept C162929932 @default.
- W3046455422 hasConcept C186450821 @default.
- W3046455422 hasConcept C202444582 @default.
- W3046455422 hasConcept C2524010 @default.
- W3046455422 hasConcept C2779593128 @default.
- W3046455422 hasConcept C33923547 @default.
- W3046455422 hasConcept C41008148 @default.
- W3046455422 hasConcept C47890412 @default.
- W3046455422 hasConcept C529865628 @default.
- W3046455422 hasConcept C70518039 @default.
- W3046455422 hasConcept C78519656 @default.
- W3046455422 hasConceptScore W3046455422C11413529 @default.
- W3046455422 hasConceptScore W3046455422C12362212 @default.
- W3046455422 hasConceptScore W3046455422C127413603 @default.
- W3046455422 hasConceptScore W3046455422C138187205 @default.
- W3046455422 hasConceptScore W3046455422C151876577 @default.
- W3046455422 hasConceptScore W3046455422C153120616 @default.
- W3046455422 hasConceptScore W3046455422C153180895 @default.
- W3046455422 hasConceptScore W3046455422C154945302 @default.
- W3046455422 hasConceptScore W3046455422C157157409 @default.
- W3046455422 hasConceptScore W3046455422C162929932 @default.
- W3046455422 hasConceptScore W3046455422C186450821 @default.
- W3046455422 hasConceptScore W3046455422C202444582 @default.
- W3046455422 hasConceptScore W3046455422C2524010 @default.
- W3046455422 hasConceptScore W3046455422C2779593128 @default.
- W3046455422 hasConceptScore W3046455422C33923547 @default.
- W3046455422 hasConceptScore W3046455422C41008148 @default.
- W3046455422 hasConceptScore W3046455422C47890412 @default.
- W3046455422 hasConceptScore W3046455422C529865628 @default.
- W3046455422 hasConceptScore W3046455422C70518039 @default.
- W3046455422 hasConceptScore W3046455422C78519656 @default.
- W3046455422 hasLocation W30464554221 @default.
- W3046455422 hasOpenAccess W3046455422 @default.
- W3046455422 hasPrimaryLocation W30464554221 @default.
- W3046455422 hasRelatedWork W1496376968 @default.
- W3046455422 hasRelatedWork W1602721037 @default.
- W3046455422 hasRelatedWork W2013051270 @default.