Abstract:Real-SR endeavors to produce high-resolution images with rich details while mitigating the impact of multiple degradation factors. Although existing methods have achieved impressive achievements in detail recovery, they still fall short when addressing regions with complex gradient arrangements due to the intensity-based linear weighting feature extraction manner. Moreover, the stochastic artifacts introduced by degradation cues during the imaging process in real LR increase the disorder of the overall image details, further complicating the perception of intrinsic gradient arrangement. To address these challenges, we innovatively introduce kernel-wise differential operations within the convolutional kernel and develop several learnable directional gradient convolutions. These convolutions are integrated in parallel with a novel linear weighting mechanism to form an Adaptive Directional Gradient Convolution (DGConv), which adaptively weights and fuses the basic directional gradients to improve the gradient arrangement perception capability for both regular and irregular textures. Coupled with DGConv, we further devise a novel equivalent parameter fusion method for DGConv that maintains its rich representational capabilities while keeping computational costs consistent with a single Vanilla Convolution (VConv), enabling DGConv to improve the performance of existing super-resolution networks without incurring additional computational expenses. To better leverage the superiority of DGConv, we further develop an Adaptive Information Interaction Block (AIIBlock) to adeptly balance the enhancement of texture and contrast while meticulously investigating the interdependencies, culminating in the creation of a DGPNet for Real-SR through simple stacking. Comparative results with 15 SOTA methods across three public datasets underscore the effectiveness and efficiency of our proposed approach.

An Improved Image Super Resolution and Its Parallel Implementation Based on CUDA.

A Fast Single-Image Super-Resolution Method Implemented with CUDA

Neural Super-Resolution in Real-Time Rendering Using Auxiliary Feature Enhancement

Study on Fast Remote Sensing Image Super-resolution Reconstruction Algorithm Based on CUDA

GPU Accelerated High-Quality Video/image Super-Resolution

Hyperspectral Image Super-Resolution Via Convolutional Neural Network

A Single-Image Super-Resolution Method Based on Progressive-Iterative Approximation

Efficient Image Super-Resolution Integration

Orientation-Aware Deep Neural Network for Real Image Super-Resolution.

Single Image Super-Resolution Using a Polymorphic Parallel CNN

A Dual CNN for Image Super-Resolution

A Deep Learning Method for Image Super-Resolution Based on Geometric Similarity.

Efficient Real-world Image Super-Resolution Via Adaptive Directional Gradient Convolution

Antialiased Super-Resolution with Parallel High-Frequency Synthesis

SINGLE IMAGE SUPER-RESOLUTION USING A NON-LOCAL 3D CONVOLUTIONAL NEURAL NETWORK

A Real-Time Super-Resolution Method Based on Convolutional Neural Networks.

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

A Novel And Effective Image Super-Resolution Reconstruction Technique Via Fast Global And Local Residual Learning Model

Super-Resolution for Noisy Images Via Deep Convolutional Neural Network

A Heterogeneous Group CNN for Image Super-Resolution

Real time super resolution reconstruction for video stream based on GPU