Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384575504> ?p ?o ?g. }
- W4384575504 endingPage "106755" @default.
- W4384575504 startingPage "106755" @default.
- W4384575504 abstract "In the past few years, convolutional neural networks (CNNs) have become the primary workhorse for image restoration tasks. However, the deficiency of modeling long-range dependencies due to the local computational property of convolution greatly limits the restoration performance. To overcome this limitation, we propose a novel multi-stage progressive convolutional Transformer to recursively restore the degraded images, termed PCformer, which enjoys a high capability for capturing local context and global dependencies with friendly computational cost. Specifically, each stage of PCformer is an asymmetric encoder–decoder network whose bottleneck is built upon a tailored Transformer block with convolution operation deployed to avoid any loss of local context. Both encoder and decoder are convolution-based modules, thus allowing to explore rich contextualized information for image recovery. Taking the low-resolution features encoded by the encoder as tokens input into the Transformer bottleneck guarantees that long-range pixel interactions are captured while reducing the computational burden. Meanwhile, we apply a gated module for filtering redundant information propagation between every two phases. In addition, long-range enhanced inpainting is further introduced to mining the ability of PCformer to exploit distant complementary features. Extensive experiments yield superior results and in particular establishing new state-of-the-art results on several image restoration tasks, including deraining (+0.37 dB on Rain13K), denoising (+0.11 dB on DND), dehazing (+0.56 dB on I-HAZE), enhancement (+0.72 dB on SICE), and shadow removal (+0.65 RMSE on ISTD). The implementation code is available at https://github.com/Jeasco/PCformer." @default.
- W4384575504 created "2023-07-18" @default.
- W4384575504 creator A5016073217 @default.
- W4384575504 creator A5018318136 @default.
- W4384575504 creator A5072912487 @default.
- W4384575504 creator A5084520562 @default.
- W4384575504 creator A5091017287 @default.
- W4384575504 date "2023-10-01" @default.
- W4384575504 modified "2023-10-17" @default.
- W4384575504 title "Progressive convolutional transformer for image restoration" @default.
- W4384575504 cites W1930824406 @default.
- W4384575504 cites W1967913888 @default.
- W4384575504 cites W1969831524 @default.
- W4384575504 cites W2045674375 @default.
- W4384575504 cites W2048695508 @default.
- W4384575504 cites W2056370875 @default.
- W4384575504 cites W2060850257 @default.
- W4384575504 cites W2110158442 @default.
- W4384575504 cites W2114770744 @default.
- W4384575504 cites W2132103241 @default.
- W4384575504 cites W2133665775 @default.
- W4384575504 cites W2209874411 @default.
- W4384575504 cites W2466666260 @default.
- W4384575504 cites W2508457857 @default.
- W4384575504 cites W2509784253 @default.
- W4384575504 cites W2556068545 @default.
- W4384575504 cites W2559264300 @default.
- W4384575504 cites W2561196672 @default.
- W4384575504 cites W2566376500 @default.
- W4384575504 cites W2603777577 @default.
- W4384575504 cites W2613155248 @default.
- W4384575504 cites W2740982616 @default.
- W4384575504 cites W2741137940 @default.
- W4384575504 cites W2752782242 @default.
- W4384575504 cites W2771617895 @default.
- W4384575504 cites W2783573276 @default.
- W4384575504 cites W2799192307 @default.
- W4384575504 cites W2866634454 @default.
- W4384575504 cites W2884068670 @default.
- W4384575504 cites W2884585870 @default.
- W4384575504 cites W2912435603 @default.
- W4384575504 cites W2913360047 @default.
- W4384575504 cites W2914992179 @default.
- W4384575504 cites W2930755307 @default.
- W4384575504 cites W2943899226 @default.
- W4384575504 cites W2962767526 @default.
- W4384575504 cites W2962782447 @default.
- W4384575504 cites W2962935103 @default.
- W4384575504 cites W2963014378 @default.
- W4384575504 cites W2963017889 @default.
- W4384575504 cites W2963073614 @default.
- W4384575504 cites W2963085671 @default.
- W4384575504 cites W2963091558 @default.
- W4384575504 cites W2963494934 @default.
- W4384575504 cites W2963800716 @default.
- W4384575504 cites W2963853051 @default.
- W4384575504 cites W2963878020 @default.
- W4384575504 cites W2964212750 @default.
- W4384575504 cites W2964221239 @default.
- W4384575504 cites W2967584026 @default.
- W4384575504 cites W2970790314 @default.
- W4384575504 cites W2971719842 @default.
- W4384575504 cites W2980047233 @default.
- W4384575504 cites W2981718299 @default.
- W4384575504 cites W2983315964 @default.
- W4384575504 cites W2989393016 @default.
- W4384575504 cites W2990007814 @default.
- W4384575504 cites W2997085666 @default.
- W4384575504 cites W2997210448 @default.
- W4384575504 cites W2998154526 @default.
- W4384575504 cites W2998249728 @default.
- W4384575504 cites W3000775737 @default.
- W4384575504 cites W3009428327 @default.
- W4384575504 cites W3034278302 @default.
- W4384575504 cites W3034504121 @default.
- W4384575504 cites W3034972231 @default.
- W4384575504 cites W3035250394 @default.
- W4384575504 cites W3035326127 @default.
- W4384575504 cites W3035731588 @default.
- W4384575504 cites W3047011367 @default.
- W4384575504 cites W3092703224 @default.
- W4384575504 cites W3095516474 @default.
- W4384575504 cites W3104725225 @default.
- W4384575504 cites W3120540810 @default.
- W4384575504 cites W3121661546 @default.
- W4384575504 cites W3158665110 @default.
- W4384575504 cites W3170697543 @default.
- W4384575504 cites W3170841864 @default.
- W4384575504 cites W3171125843 @default.
- W4384575504 cites W3173269149 @default.
- W4384575504 cites W3174300208 @default.
- W4384575504 cites W3176096490 @default.
- W4384575504 cites W3177127346 @default.
- W4384575504 cites W3186182384 @default.
- W4384575504 cites W3202362072 @default.
- W4384575504 cites W3207649350 @default.
- W4384575504 cites W3207918547 @default.
- W4384575504 cites W4225672218 @default.