Search Results for author:
Found 10 papers, 9 papers with code
Variational Deep Image Restoration
This paper presents a new variational inference framework for image restoration and a convolutional neural network (CNN) structure that can solve the restoration problems described by the proposed framework.
Training Patch Analysis and Mining Skills for Image Restoration Deep Neural Networks
Eventually, we propose a guideline for the patch extraction from given training images.
A Dynamic Residual Self-Attention Network for Lightweight Single Image Super-Resolution
We also propose a residual self-attention (RSA) module to further boost the performance, which produces 3-dimensional attention maps without additional parameters by cooperating with residual structures.
Variational Deep Image Denoising
These methods separate the original problem into easier sub-problems and thus have shown improved performance than the naively trained CNN.
Deep Universal Blind Image Denoising
Traditionally, many researchers have investigated image priors for the denoising, within the Bayesian perspective based on image properties and statistics.
Meta-Transfer Learning for Zero-Shot Super-Resolution
Despite their remarkable performance based on the external dataset, they cannot exploit internal information within a specific image.
Transfer Learning from Synthetic to Real-Noise Denoising with Adaptive Instance Normalization
Real-noise denoising is a challenging task because the statistics of real-noise do not follow the normal distribution, and they are also spatially and temporally changing.
Ranked #12 on
Image Denoising
on DND
(using extra training data)
Natural and Realistic Single Image Super-Resolution with Explicit Natural Manifold Discrimination
Recently, many convolutional neural networks for single image super-resolution (SISR) have been proposed, which focus on reconstructing the high-resolution images in terms of objective distortion measures.
Handwritten Text Segmentation via End-to-End Learning of Convolutional Neural Network
For training our network, we develop a cross-entropy based loss function that addresses the imbalance problems.
Joint High Dynamic Range Imaging and Super-Resolution from a Single Image
This paper presents a new framework for jointly enhancing the resolution and the dynamic range of an image, i. e., simultaneous super-resolution (SR) and high dynamic range imaging (HDRI), based on a convolutional neural network (CNN).
