Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384207671> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4384207671 endingPage "109819" @default.
- W4384207671 startingPage "109819" @default.
- W4384207671 abstract "Downsampling operations such as max pooling or strided convolution are ubiquitously utilized in Convolutional Neural Networks (CNNs) to aggregate local features, enlarge receptive field, and minimize computational overhead. However, for a semantic segmentation task, pooling features over the local neighbourhood may result in the loss of important spatial information, which is conducive for pixel-wise predictions. To address this issue, we introduce a simple yet effective pooling operation called the Haar Wavelet-based Downsampling (HWD) module. This module can be easily integrated into CNNs to enhance the performance of semantic segmentation models. The core idea of HWD is to apply Haar wavelet transform for reducing the spatial resolution of feature maps while preserving as much information as possible. Furthermore, to investigate the benefits of HWD, we propose a novel metric, named as feature entropy index (FEI), which measures the degree of information uncertainty after downsampling in CNNs. Specifically, the FEI can be used to indicate the ability of downsampling methods to preserve essential information in semantic segmentation. Our comprehensive experiments demonstrate that the proposed HWD module could (1) effectively improve the segmentation performance across different modality image datasets with various CNN architectures, and (2) efficiently reduce information uncertainty compared to the conventional downsampling methods. Our implementation are available at https://github.com/apple1986/HWD." @default.
- W4384207671 created "2023-07-14" @default.
- W4384207671 creator A5026390334 @default.
- W4384207671 creator A5033484584 @default.
- W4384207671 creator A5034757748 @default.
- W4384207671 creator A5037246170 @default.
- W4384207671 creator A5067358905 @default.
- W4384207671 creator A5085294808 @default.
- W4384207671 date "2023-11-01" @default.
- W4384207671 modified "2023-09-26" @default.
- W4384207671 title "Haar wavelet downsampling: A simple but effective downsampling module for semantic segmentation" @default.
- W4384207671 cites W2016277672 @default.
- W4384207671 cites W2049016464 @default.
- W4384207671 cites W2097498222 @default.
- W4384207671 cites W2101946599 @default.
- W4384207671 cites W2105902929 @default.
- W4384207671 cites W2114582993 @default.
- W4384207671 cites W2133665775 @default.
- W4384207671 cites W2141983208 @default.
- W4384207671 cites W2921526458 @default.
- W4384207671 cites W2949687910 @default.
- W4384207671 cites W2963881378 @default.
- W4384207671 cites W2994114312 @default.
- W4384207671 cites W3014641072 @default.
- W4384207671 cites W3017123215 @default.
- W4384207671 cites W3085128889 @default.
- W4384207671 cites W3106651317 @default.
- W4384207671 cites W3113958627 @default.
- W4384207671 cites W3116109061 @default.
- W4384207671 cites W3132941258 @default.
- W4384207671 cites W3161864725 @default.
- W4384207671 cites W3181578532 @default.
- W4384207671 cites W3199550750 @default.
- W4384207671 cites W3201484072 @default.
- W4384207671 cites W3217020749 @default.
- W4384207671 cites W4205711607 @default.
- W4384207671 cites W4212858574 @default.
- W4384207671 cites W4284892523 @default.
- W4384207671 cites W4285308517 @default.
- W4384207671 cites W4290983194 @default.
- W4384207671 cites W4313479120 @default.
- W4384207671 doi "https://doi.org/10.1016/j.patcog.2023.109819" @default.
- W4384207671 hasPublicationYear "2023" @default.
- W4384207671 type Work @default.
- W4384207671 citedByCount "0" @default.
- W4384207671 crossrefType "journal-article" @default.
- W4384207671 hasAuthorship W4384207671A5026390334 @default.
- W4384207671 hasAuthorship W4384207671A5033484584 @default.
- W4384207671 hasAuthorship W4384207671A5034757748 @default.
- W4384207671 hasAuthorship W4384207671A5037246170 @default.
- W4384207671 hasAuthorship W4384207671A5067358905 @default.
- W4384207671 hasAuthorship W4384207671A5085294808 @default.
- W4384207671 hasConcept C110384440 @default.
- W4384207671 hasConcept C115961682 @default.
- W4384207671 hasConcept C153180895 @default.
- W4384207671 hasConcept C154945302 @default.
- W4384207671 hasConcept C196216189 @default.
- W4384207671 hasConcept C2780423554 @default.
- W4384207671 hasConcept C41008148 @default.
- W4384207671 hasConcept C46286280 @default.
- W4384207671 hasConcept C47432892 @default.
- W4384207671 hasConcept C70437156 @default.
- W4384207671 hasConcept C81363708 @default.
- W4384207671 hasConcept C89600930 @default.
- W4384207671 hasConceptScore W4384207671C110384440 @default.
- W4384207671 hasConceptScore W4384207671C115961682 @default.
- W4384207671 hasConceptScore W4384207671C153180895 @default.
- W4384207671 hasConceptScore W4384207671C154945302 @default.
- W4384207671 hasConceptScore W4384207671C196216189 @default.
- W4384207671 hasConceptScore W4384207671C2780423554 @default.
- W4384207671 hasConceptScore W4384207671C41008148 @default.
- W4384207671 hasConceptScore W4384207671C46286280 @default.
- W4384207671 hasConceptScore W4384207671C47432892 @default.
- W4384207671 hasConceptScore W4384207671C70437156 @default.
- W4384207671 hasConceptScore W4384207671C81363708 @default.
- W4384207671 hasConceptScore W4384207671C89600930 @default.
- W4384207671 hasLocation W43842076711 @default.
- W4384207671 hasOpenAccess W4384207671 @default.
- W4384207671 hasPrimaryLocation W43842076711 @default.
- W4384207671 hasRelatedWork W1577789985 @default.
- W4384207671 hasRelatedWork W2090493506 @default.
- W4384207671 hasRelatedWork W2138087993 @default.
- W4384207671 hasRelatedWork W2158234487 @default.
- W4384207671 hasRelatedWork W2545237249 @default.
- W4384207671 hasRelatedWork W2905787433 @default.
- W4384207671 hasRelatedWork W3132543298 @default.
- W4384207671 hasRelatedWork W4206491577 @default.
- W4384207671 hasRelatedWork W634117680 @default.
- W4384207671 hasRelatedWork W191391683 @default.
- W4384207671 hasVolume "143" @default.
- W4384207671 isParatext "false" @default.
- W4384207671 isRetracted "false" @default.
- W4384207671 workType "article" @default.