Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951081618> ?p ?o ?g. }
- W2951081618 abstract "Finding semantic correspondences is a challenging problem. With the breakthrough of CNNs stronger features are available for tasks like classification but not specifically for the requirements of semantic matching. In the following we present a weakly supervised metric learning approach which generates stronger features by encoding far more context than previous methods. First, we generate more suitable training data using a geometrically informed correspondence mining method which is less prone to spurious matches and requires only image category labels as supervision. Second, we introduce a new convolutional layer which is a learned mixture of differently strided convolutions and allows the network to encode implicitly more context while preserving matching accuracy at the same time. The strong geometric encoding on the feature side enables us to learn a semantic flow network, which generates more natural deformations than parametric transformation based models and is able to jointly predict foreground regions at the same time. Our semantic flow network outperforms current state-of-the-art on several semantic matching benchmarks and the learned features show astonishing performance regarding simple nearest neighbor matching." @default.
- W2951081618 created "2019-06-27" @default.
- W2951081618 creator A5008066954 @default.
- W2951081618 creator A5013847587 @default.
- W2951081618 creator A5028087949 @default.
- W2951081618 creator A5028405919 @default.
- W2951081618 creator A5084415727 @default.
- W2951081618 date "2019-06-17" @default.
- W2951081618 modified "2023-09-27" @default.
- W2951081618 title "Multi-Scale Convolutions for Learning Context Aware Feature Representations" @default.
- W2951081618 cites W1686810756 @default.
- W2951081618 cites W1849277567 @default.
- W2951081618 cites W1926639317 @default.
- W2951081618 cites W1960289438 @default.
- W2951081618 cites W1984034752 @default.
- W2951081618 cites W2090518410 @default.
- W2951081618 cites W2104408738 @default.
- W2951081618 cites W2108598243 @default.
- W2951081618 cites W2115733720 @default.
- W2951081618 cites W2124861766 @default.
- W2951081618 cites W2126080861 @default.
- W2951081618 cites W2128409098 @default.
- W2951081618 cites W2137278143 @default.
- W2951081618 cites W2151103935 @default.
- W2951081618 cites W2157364932 @default.
- W2951081618 cites W2161236525 @default.
- W2951081618 cites W2161969291 @default.
- W2951081618 cites W2163605009 @default.
- W2951081618 cites W2164918853 @default.
- W2951081618 cites W2412002662 @default.
- W2951081618 cites W2464606141 @default.
- W2951081618 cites W2554120895 @default.
- W2951081618 cites W2556967412 @default.
- W2951081618 cites W2566812041 @default.
- W2951081618 cites W2593948489 @default.
- W2951081618 cites W2601564443 @default.
- W2951081618 cites W2604233003 @default.
- W2951081618 cites W2747550417 @default.
- W2951081618 cites W2886782161 @default.
- W2951081618 cites W2963020784 @default.
- W2951081618 cites W2963325280 @default.
- W2951081618 cites W2963840672 @default.
- W2951081618 cites W2964141676 @default.
- W2951081618 cites W2964188292 @default.
- W2951081618 cites W2964213755 @default.
- W2951081618 cites W603908379 @default.
- W2951081618 cites W764651262 @default.
- W2951081618 cites W2435623039 @default.
- W2951081618 hasPublicationYear "2019" @default.
- W2951081618 type Work @default.
- W2951081618 sameAs 2951081618 @default.
- W2951081618 citedByCount "0" @default.
- W2951081618 crossrefType "posted-content" @default.
- W2951081618 hasAuthorship W2951081618A5008066954 @default.
- W2951081618 hasAuthorship W2951081618A5013847587 @default.
- W2951081618 hasAuthorship W2951081618A5028087949 @default.
- W2951081618 hasAuthorship W2951081618A5028405919 @default.
- W2951081618 hasAuthorship W2951081618A5084415727 @default.
- W2951081618 hasConcept C104317684 @default.
- W2951081618 hasConcept C105795698 @default.
- W2951081618 hasConcept C117251300 @default.
- W2951081618 hasConcept C119857082 @default.
- W2951081618 hasConcept C125411270 @default.
- W2951081618 hasConcept C138885662 @default.
- W2951081618 hasConcept C151730666 @default.
- W2951081618 hasConcept C153180895 @default.
- W2951081618 hasConcept C154945302 @default.
- W2951081618 hasConcept C162324750 @default.
- W2951081618 hasConcept C165064840 @default.
- W2951081618 hasConcept C176217482 @default.
- W2951081618 hasConcept C185592680 @default.
- W2951081618 hasConcept C204241405 @default.
- W2951081618 hasConcept C21547014 @default.
- W2951081618 hasConcept C2776401178 @default.
- W2951081618 hasConcept C2779343474 @default.
- W2951081618 hasConcept C33923547 @default.
- W2951081618 hasConcept C41008148 @default.
- W2951081618 hasConcept C41895202 @default.
- W2951081618 hasConcept C45347329 @default.
- W2951081618 hasConcept C50644808 @default.
- W2951081618 hasConcept C55493867 @default.
- W2951081618 hasConcept C66746571 @default.
- W2951081618 hasConcept C86803240 @default.
- W2951081618 hasConcept C97256817 @default.
- W2951081618 hasConceptScore W2951081618C104317684 @default.
- W2951081618 hasConceptScore W2951081618C105795698 @default.
- W2951081618 hasConceptScore W2951081618C117251300 @default.
- W2951081618 hasConceptScore W2951081618C119857082 @default.
- W2951081618 hasConceptScore W2951081618C125411270 @default.
- W2951081618 hasConceptScore W2951081618C138885662 @default.
- W2951081618 hasConceptScore W2951081618C151730666 @default.
- W2951081618 hasConceptScore W2951081618C153180895 @default.
- W2951081618 hasConceptScore W2951081618C154945302 @default.
- W2951081618 hasConceptScore W2951081618C162324750 @default.
- W2951081618 hasConceptScore W2951081618C165064840 @default.
- W2951081618 hasConceptScore W2951081618C176217482 @default.
- W2951081618 hasConceptScore W2951081618C185592680 @default.
- W2951081618 hasConceptScore W2951081618C204241405 @default.
- W2951081618 hasConceptScore W2951081618C21547014 @default.
- W2951081618 hasConceptScore W2951081618C2776401178 @default.