2024/6/9 -SRCNN. Super-resolution Convolutional Neural Network for image upscaling. Based on Image Super-Resolution Using Deep Convolutional Networks. The model is ...
2024/6/18 -In this study, the super-resolution convolutional neural network (SRCNN) scheme, which is the emerging deep-learning-based super-resolution method for ...
2024/6/15 -We compare the proposed BSRN with state-of-the-art lightweight SR approaches, including SRCNN, FSRCNN, LapSRN, VDSR, DRCN, SRDenseNet, CARN-M, CARN, DRRN, ...
2024/6/29 -SRCNN has achieved notable outcomes in reconstruction of the medical CT images (Umehara et al., 2018) and digital rock images (Bai and Berezovsky, 2020 ...
2日前 -For Single Image Super-Resolution (SISR) tasks, SRCNN [13] is the pioneer of applying DNN to image super-resolution. Then, followed by FSRCNN [14] and ESPCN ...
4時間前 -Dong et al. proposed SRCNN (image super-resolution using deep convolutional network, SRCNN) for the first time to achieve end-to-end image SR ...
2024/6/11 -With SRCNN, the receptive field can be increased by using more convolutional layers to improve the results; however, gradient vanishing/exploding and ...
2024/6/17 -FSRCNN [10] proposed a network paradigm that places the upsampling step in the last stage, replacing the enormous computational cost incurred by SRCNN [9] which ...
2024/6/20 -The foundation of SISR lies in the utilization of deep neural networks, including but not limited to architectures like VDSR, SRCNN, and EDSR, which have ...
2024/6/23 -The pioneering work of SRCNN [27] marked the initiation of applying deep learning to SISR, employing a three-layer convolutional neural network for the task.