Matches in SemOpenAlex for { <https://semopenalex.org/work/W4289598265> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4289598265 abstract "Convolutional neural networks (CNNs) have been the de facto standard in a diverse set of computer vision tasks for many years. Especially, deep neural networks based on seminal architectures such as U-shaped models with skip-connections or atrous convolution with pyramid pooling have been tailored to a wide range of medical image analysis tasks. The main advantage of such architectures is that they are prone to detaining versatile local features. However, as a general consensus, CNNs fail to capture long-range dependencies and spatial correlations due to the intrinsic property of confined receptive field size of convolution operations. Alternatively, Transformer, profiting from global information modelling that stems from the self-attention mechanism, has recently attained remarkable performance in natural language processing and computer vision. Nevertheless, previous studies prove that both local and global features are critical for a deep model in dense prediction, such as segmenting complicated structures with disparate shapes and configurations. To this end, this paper proposes TransDeepLab, a novel DeepLab-like pure Transformer for medical image segmentation. Specifically, we exploit hierarchical Swin-Transformer with shifted windows to extend the DeepLabv3 and model the Atrous Spatial Pyramid Pooling (ASPP) module. A thorough search of the relevant literature yielded that we are the first to model the seminal DeepLab model with a pure Transformer-based model. Extensive experiments on various medical image segmentation tasks verify that our approach performs superior or on par with most contemporary works on an amalgamation of Vision Transformer and CNN-based methods, along with a significant reduction of model complexity. The codes and trained models are publicly available at https://github.com/rezazad68/transdeeplab" @default.
- W4289598265 created "2022-08-03" @default.
- W4289598265 creator A5000174202 @default.
- W4289598265 creator A5008523578 @default.
- W4289598265 creator A5015355317 @default.
- W4289598265 creator A5015411150 @default.
- W4289598265 creator A5019023164 @default.
- W4289598265 creator A5064747056 @default.
- W4289598265 creator A5087512747 @default.
- W4289598265 date "2022-08-01" @default.
- W4289598265 modified "2023-10-14" @default.
- W4289598265 title "TransDeepLab: Convolution-Free Transformer-based DeepLab v3+ for Medical Image Segmentation" @default.
- W4289598265 doi "https://doi.org/10.48550/arxiv.2208.00713" @default.
- W4289598265 hasPublicationYear "2022" @default.
- W4289598265 type Work @default.
- W4289598265 citedByCount "0" @default.
- W4289598265 crossrefType "posted-content" @default.
- W4289598265 hasAuthorship W4289598265A5000174202 @default.
- W4289598265 hasAuthorship W4289598265A5008523578 @default.
- W4289598265 hasAuthorship W4289598265A5015355317 @default.
- W4289598265 hasAuthorship W4289598265A5015411150 @default.
- W4289598265 hasAuthorship W4289598265A5019023164 @default.
- W4289598265 hasAuthorship W4289598265A5064747056 @default.
- W4289598265 hasAuthorship W4289598265A5087512747 @default.
- W4289598265 hasBestOaLocation W42895982651 @default.
- W4289598265 hasConcept C121332964 @default.
- W4289598265 hasConcept C124504099 @default.
- W4289598265 hasConcept C153180895 @default.
- W4289598265 hasConcept C154945302 @default.
- W4289598265 hasConcept C165801399 @default.
- W4289598265 hasConcept C31972630 @default.
- W4289598265 hasConcept C41008148 @default.
- W4289598265 hasConcept C62520636 @default.
- W4289598265 hasConcept C66322947 @default.
- W4289598265 hasConcept C70437156 @default.
- W4289598265 hasConcept C81363708 @default.
- W4289598265 hasConcept C89600930 @default.
- W4289598265 hasConceptScore W4289598265C121332964 @default.
- W4289598265 hasConceptScore W4289598265C124504099 @default.
- W4289598265 hasConceptScore W4289598265C153180895 @default.
- W4289598265 hasConceptScore W4289598265C154945302 @default.
- W4289598265 hasConceptScore W4289598265C165801399 @default.
- W4289598265 hasConceptScore W4289598265C31972630 @default.
- W4289598265 hasConceptScore W4289598265C41008148 @default.
- W4289598265 hasConceptScore W4289598265C62520636 @default.
- W4289598265 hasConceptScore W4289598265C66322947 @default.
- W4289598265 hasConceptScore W4289598265C70437156 @default.
- W4289598265 hasConceptScore W4289598265C81363708 @default.
- W4289598265 hasConceptScore W4289598265C89600930 @default.
- W4289598265 hasLocation W42895982651 @default.
- W4289598265 hasOpenAccess W4289598265 @default.
- W4289598265 hasPrimaryLocation W42895982651 @default.
- W4289598265 hasRelatedWork W11130107 @default.
- W4289598265 hasRelatedWork W11270157 @default.
- W4289598265 hasRelatedWork W12868778 @default.
- W4289598265 hasRelatedWork W14128562 @default.
- W4289598265 hasRelatedWork W1602910 @default.
- W4289598265 hasRelatedWork W1679810 @default.
- W4289598265 hasRelatedWork W2526871 @default.
- W4289598265 hasRelatedWork W274842 @default.
- W4289598265 hasRelatedWork W7789328 @default.
- W4289598265 hasRelatedWork W3000238 @default.
- W4289598265 isParatext "false" @default.
- W4289598265 isRetracted "false" @default.
- W4289598265 workType "article" @default.