Matches in SemOpenAlex for { <https://semopenalex.org/work/W2572875100> ?p ?o ?g. }
- W2572875100 abstract "Deep Convolution Neural Networks (DCNNs) are capable of learning unprecedentedly effective image representations. However, their ability in handling significant local and global image rotations remains limited. In this paper, we propose Active Rotating Filters (ARFs) that actively rotate during convolution and produce feature maps with location and orientation explicitly encoded. An ARF acts as a virtual filter bank containing the filter itself and its multiple unmaterialised rotated versions. During back-propagation, an ARF is collectively updated using errors from all its rotated versions. DCNNs using ARFs, referred to as Oriented Response Networks (ORNs), can produce within-class rotation-invariant deep features while maintaining inter-class discrimination for classification tasks. The oriented response produced by ORNs can also be used for image and object orientation estimation tasks. Over multiple state-of-the-art DCNN architectures, such as VGG, ResNet, and STN, we consistently observe that replacing regular filters with the proposed ARFs leads to significant reduction in the number of network parameters and improvement in classification performance. We report the best results on several commonly used benchmarks." @default.
- W2572875100 created "2017-01-26" @default.
- W2572875100 creator A5015317495 @default.
- W2572875100 creator A5049689760 @default.
- W2572875100 creator A5066032427 @default.
- W2572875100 creator A5086749611 @default.
- W2572875100 date "2017-01-07" @default.
- W2572875100 modified "2023-09-23" @default.
- W2572875100 title "Oriented Response Networks" @default.
- W2572875100 cites W1523493493 @default.
- W2572875100 cites W1686810756 @default.
- W2572875100 cites W1849277567 @default.
- W2572875100 cites W1975868529 @default.
- W2572875100 cites W1994197834 @default.
- W2572875100 cites W2008213480 @default.
- W2572875100 cites W2024986448 @default.
- W2572875100 cites W2025183033 @default.
- W2572875100 cites W2072072671 @default.
- W2572875100 cites W2099471712 @default.
- W2572875100 cites W2108598243 @default.
- W2572875100 cites W2124386111 @default.
- W2572875100 cites W2130258210 @default.
- W2572875100 cites W2156116778 @default.
- W2572875100 cites W2161969291 @default.
- W2572875100 cites W2162018804 @default.
- W2572875100 cites W2163352848 @default.
- W2572875100 cites W2163808566 @default.
- W2572875100 cites W2164294799 @default.
- W2572875100 cites W2167383966 @default.
- W2572875100 cites W2168026118 @default.
- W2572875100 cites W2173520492 @default.
- W2572875100 cites W2176412452 @default.
- W2572875100 cites W2220894487 @default.
- W2572875100 cites W2243462087 @default.
- W2572875100 cites W2302255633 @default.
- W2572875100 cites W2340427832 @default.
- W2572875100 cites W2401231614 @default.
- W2572875100 cites W2461725797 @default.
- W2572875100 cites W2483076098 @default.
- W2572875100 cites W2949650786 @default.
- W2572875100 cites W2950328304 @default.
- W2572875100 cites W2950831665 @default.
- W2572875100 cites W2952054889 @default.
- W2572875100 cites W2952390042 @default.
- W2572875100 cites W3038058348 @default.
- W2572875100 cites W3118608800 @default.
- W2572875100 cites W603908379 @default.
- W2572875100 cites W6908809 @default.
- W2572875100 doi "https://doi.org/10.48550/arxiv.1701.01833" @default.
- W2572875100 hasPublicationYear "2017" @default.
- W2572875100 type Work @default.
- W2572875100 sameAs 2572875100 @default.
- W2572875100 citedByCount "23" @default.
- W2572875100 countsByYear W25728751002017 @default.
- W2572875100 countsByYear W25728751002018 @default.
- W2572875100 countsByYear W25728751002019 @default.
- W2572875100 countsByYear W25728751002020 @default.
- W2572875100 crossrefType "posted-content" @default.
- W2572875100 hasAuthorship W2572875100A5015317495 @default.
- W2572875100 hasAuthorship W2572875100A5049689760 @default.
- W2572875100 hasAuthorship W2572875100A5066032427 @default.
- W2572875100 hasAuthorship W2572875100A5086749611 @default.
- W2572875100 hasBestOaLocation W25728751001 @default.
- W2572875100 hasConcept C100515483 @default.
- W2572875100 hasConcept C106131492 @default.
- W2572875100 hasConcept C108583219 @default.
- W2572875100 hasConcept C138885662 @default.
- W2572875100 hasConcept C153180895 @default.
- W2572875100 hasConcept C154945302 @default.
- W2572875100 hasConcept C16345878 @default.
- W2572875100 hasConcept C190470478 @default.
- W2572875100 hasConcept C2524010 @default.
- W2572875100 hasConcept C2776401178 @default.
- W2572875100 hasConcept C31972630 @default.
- W2572875100 hasConcept C33923547 @default.
- W2572875100 hasConcept C37914503 @default.
- W2572875100 hasConcept C41008148 @default.
- W2572875100 hasConcept C41895202 @default.
- W2572875100 hasConcept C45347329 @default.
- W2572875100 hasConcept C50644808 @default.
- W2572875100 hasConcept C74050887 @default.
- W2572875100 hasConcept C81363708 @default.
- W2572875100 hasConceptScore W2572875100C100515483 @default.
- W2572875100 hasConceptScore W2572875100C106131492 @default.
- W2572875100 hasConceptScore W2572875100C108583219 @default.
- W2572875100 hasConceptScore W2572875100C138885662 @default.
- W2572875100 hasConceptScore W2572875100C153180895 @default.
- W2572875100 hasConceptScore W2572875100C154945302 @default.
- W2572875100 hasConceptScore W2572875100C16345878 @default.
- W2572875100 hasConceptScore W2572875100C190470478 @default.
- W2572875100 hasConceptScore W2572875100C2524010 @default.
- W2572875100 hasConceptScore W2572875100C2776401178 @default.
- W2572875100 hasConceptScore W2572875100C31972630 @default.
- W2572875100 hasConceptScore W2572875100C33923547 @default.
- W2572875100 hasConceptScore W2572875100C37914503 @default.
- W2572875100 hasConceptScore W2572875100C41008148 @default.
- W2572875100 hasConceptScore W2572875100C41895202 @default.
- W2572875100 hasConceptScore W2572875100C45347329 @default.
- W2572875100 hasConceptScore W2572875100C50644808 @default.
- W2572875100 hasConceptScore W2572875100C74050887 @default.