Rethinking image super resolution from long-tailed distribution learning perspective

Y Gou, P Hu, J Lv, H Zhu… - Proceedings of the IEEE …, 2023 - openaccess.thecvf.com
Existing studies have empirically observed that the resolution of the low-frequency region is
easier to enhance than that of the high-frequency one. Although plentiful works have been …

Combination of super-resolution reconstruction and SGA-Net for marsh vegetation map** using multi-resolution multispectral and hyperspectral images

B Fu, X Sun, Y Li, Z Lao, T Deng, H He… - … Journal of Digital …, 2023 - Taylor & Francis
Vegetation is crucial for wetland ecosystems. Human activities and climate changes are
increasingly threatening wetland ecosystems. Combining satellite images and deep …

Learning re-sampling methods with parameter attribution for image super-resolution

X Luo, Y **e, Y Qu - Advances in Neural Information …, 2023 - proceedings.neurips.cc
Single image super-resolution (SISR) has made a significant breakthrough benefiting from
the prevalent rise of deep neural networks and large-scale training samples. The …

Data augmentation for multi-image super-resolution

M Ziaja, J Nalepa, M Kawulok - IGARSS 2022-2022 IEEE …, 2022 - ieeexplore.ieee.org
Super-resolution reconstruction consists in generating a high-resolution image from a single
low-resolution image or multiple images presenting the same area of interest. Existing state …

RepCaM: Re-parameterization Content-aware Modulation for Neural Video Delivery

R Zhang, L Du, J Liu, C Song, F Wang, X Li… - Proceedings of the 33rd …, 2023 - dl.acm.org
Recently, content-aware methods have been utilized to reduce the bandwidth and improve
the quality of Internet video delivery. Existing methods train corresponding content-aware …

DDA: A dynamic difficulty-aware data augmenter for image super-resolution

X Zhang, T Dai, B Chen, ST **a - 2023 International Joint …, 2023 - ieeexplore.ieee.org
Deep neural networks (DNNs) have been recently widely used in image super-resolution
(SR) and have achieved remarkable performance. However, most existing methods focus on …

Rethinking Imbalance in Image Super-Resolution for Efficient Inference

W Yu, B Yang, Q Liu, J Li, S Zhang, X Ji - The Thirty-eighth Annual … - openreview.net
Existing super-resolution (SR) methods optimize all model weights equally using $\mathcal
{L} _1 $ or $\mathcal {L} _2 $ losses by uniformly sampling image patches without …