FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4316660708> ?p ?o ?g. }

• W4316660708 endingPage "15" @default.
• W4316660708 startingPage "1" @default.
• W4316660708 abstract "Image deblurring is a representative low-level vision task that aims to estimate latent sharp images from blurred images. Recently, convolutional neural network (CNN)-based methods have dominated image deblurring. However, traditional CNN-based deblurring methods suffer from two essential issues: first, existing multiscale deblurring methods process blurred images at different scales through sub-networks with the same composition, which limits the model performance. Second, the convolutional layers fail to adapt to the input content and cannot effectively capture long-range dependencies. To alleviate the above issues, we rethink the multiscale architecture that follows a coarse-to-fine strategy and propose a novel hybrid architecture that combines CNN and transformer (CTMS). CTMS has three distinct features. First, the finer-scale sub-networks in CTMS are designed as architectures with larger receptive fields to obtain the pixel values around the blur, which can be used to efficiently handle large-area blur. Then, we propose a feature modulation network to alleviate the disadvantages of CNN sub-networks that lack input content adaptation. Finally, we design an efficient transformer block, which significantly reduces the computational burden and requires no pre-training. Our proposed deblurring model is extensively evaluated on several benchmark datasets, and achieves superior performance compared to state-of-the-art deblurring methods. Especially, the peak signal to noise ratio (PSNR) and structural similarity (SSIM) values are 32.73 dB and 0.959, respectively, on the popular dataset GoPro. In addition, we conduct joint evaluation experiments on the proposed method deblurring performance, object detection, and image segmentation to demonstrate the effectiveness of CTMS for subsequent high-level computer vision tasks." @default.
• W4316660708 created "2023-01-17" @default.
• W4316660708 creator A5013308588 @default.
• W4316660708 creator A5038174569 @default.
• W4316660708 creator A5067650828 @default.
• W4316660708 creator A5069672411 @default.
• W4316660708 date "2023-01-01" @default.
• W4316660708 modified "2023-10-16" @default.
• W4316660708 title "Rethinking Image Deblurring via CNN-Transformer Multiscale Hybrid Architecture" @default.
• W4316660708 cites W1580389772 @default.
• W4316660708 cites W1916935112 @default.
• W4316660708 cites W2046119925 @default.
• W4316660708 cites W2124500511 @default.
• W4316660708 cites W2167307343 @default.
• W4316660708 cites W2194775991 @default.
• W4316660708 cites W233979554 @default.
• W4316660708 cites W2560023338 @default.
• W4316660708 cites W2560533888 @default.
• W4316660708 cites W2564023417 @default.
• W4316660708 cites W2573798160 @default.
• W4316660708 cites W2607041014 @default.
• W4316660708 cites W2759428153 @default.
• W4316660708 cites W2765507148 @default.
• W4316660708 cites W2806224701 @default.
• W4316660708 cites W2961218591 @default.
• W4316660708 cites W2962785568 @default.
• W4316660708 cites W2963312584 @default.
• W4316660708 cites W2963814095 @default.
• W4316660708 cites W2964030969 @default.
• W4316660708 cites W2965217508 @default.
• W4316660708 cites W2966284335 @default.
• W4316660708 cites W2975950300 @default.
• W4316660708 cites W2982795046 @default.
• W4316660708 cites W2988641380 @default.
• W4316660708 cites W2998249728 @default.
• W4316660708 cites W3013529009 @default.
• W4316660708 cites W3017136408 @default.
• W4316660708 cites W3034789174 @default.
• W4316660708 cites W3035484352 @default.
• W4316660708 cites W3040557872 @default.
• W4316660708 cites W3085824656 @default.
• W4316660708 cites W3097463671 @default.
• W4316660708 cites W3097762837 @default.
• W4316660708 cites W3105070560 @default.
• W4316660708 cites W3106135392 @default.
• W4316660708 cites W3120680448 @default.
• W4316660708 cites W3131500599 @default.
• W4316660708 cites W3138516171 @default.
• W4316660708 cites W3155350616 @default.
• W4316660708 cites W3170697543 @default.
• W4316660708 cites W3171125843 @default.
• W4316660708 cites W3174756006 @default.
• W4316660708 cites W3193296307 @default.
• W4316660708 cites W3200683480 @default.
• W4316660708 cites W3202040256 @default.
• W4316660708 cites W3203848400 @default.
• W4316660708 cites W3207516177 @default.
• W4316660708 cites W3210068417 @default.
• W4316660708 cites W3212022416 @default.
• W4316660708 cites W3213236098 @default.
• W4316660708 cites W4213019189 @default.
• W4316660708 cites W4220974521 @default.
• W4316660708 cites W4225672218 @default.
• W4316660708 cites W4226043703 @default.
• W4316660708 cites W4282575541 @default.
• W4316660708 cites W4285105199 @default.
• W4316660708 cites W4285105754 @default.
• W4316660708 cites W4288700147 @default.
• W4316660708 cites W4312812783 @default.
• W4316660708 cites W4312938066 @default.
• W4316660708 doi "https://doi.org/10.1109/tim.2022.3230482" @default.
• W4316660708 hasPublicationYear "2023" @default.
• W4316660708 type Work @default.
• W4316660708 citedByCount "0" @default.
• W4316660708 crossrefType "journal-article" @default.
• W4316660708 hasAuthorship W4316660708A5013308588 @default.
• W4316660708 hasAuthorship W4316660708A5038174569 @default.
• W4316660708 hasAuthorship W4316660708A5067650828 @default.
• W4316660708 hasAuthorship W4316660708A5069672411 @default.
• W4316660708 hasConcept C106430172 @default.
• W4316660708 hasConcept C108583219 @default.
• W4316660708 hasConcept C115961682 @default.
• W4316660708 hasConcept C124504099 @default.
• W4316660708 hasConcept C13280743 @default.
• W4316660708 hasConcept C153180895 @default.
• W4316660708 hasConcept C154945302 @default.
• W4316660708 hasConcept C185798385 @default.
• W4316660708 hasConcept C205649164 @default.
• W4316660708 hasConcept C2777693668 @default.
• W4316660708 hasConcept C31972630 @default.
• W4316660708 hasConcept C41008148 @default.
• W4316660708 hasConcept C81363708 @default.
• W4316660708 hasConcept C89600930 @default.
• W4316660708 hasConcept C9417928 @default.
• W4316660708 hasConceptScore W4316660708C106430172 @default.
• W4316660708 hasConceptScore W4316660708C108583219 @default.
• W4316660708 hasConceptScore W4316660708C115961682 @default.
• W4316660708 hasConceptScore W4316660708C124504099 @default.

• next