【图像去噪】论文精读:Rotation-Equivariant Self-Supervised Method in Image Denoising(AdaReNet)

请先看【专栏介绍文章】:【图像去噪(Image Denoising)】关于【图像去噪】专栏的相关说明,包含适配人群、专栏简介、专栏亮点、阅读方法、定价理由、品质承诺、关于更新、去噪概述、文章目录、资料汇总、问题汇总(更新中)

文章目录

  • 前言
  • Abstract
  • 1. Introduction
  • 2. Related Work and Prior Knowledge
    • 2.1. Image Denoising
    • 2.2. Rotation Equivariance
    • 2.3. Prior Knowledge about Equivariance
  • 3. Proposed Method
    • 3.1. ECNNs for Self-supervised Image Denoising
      • 3.1.1. Notations and Concepts
      • 3.1.2. Equivariance of Downsampling and Upsampling
      • 3.1.3. Analysis of Complete U-Net Network
    • 3.2. Proposed Adaptive Network AdaReNet
  • 4. Experiments
    • 4.1. Experiments on Compared Methods
    • 4.2. Ablation Study
  • 5. Conclusion


前言

论文题目:Rotation-Equivariant Self-Supervised Method in Image Denoising —— 图像去噪中的旋转等变自监督方法

论文地址:Rotation-Equivariant Self-Supervised Method in Image Denoising

论文源码:https://github.com/liuhanze623/AdaReNet

CVPR 2025!

Abstract

近年来,自监督图像去噪方法引起了广泛的研

你可能感兴趣的:(深度学习,人工智能,图像处理,计算机视觉,图像去噪,论文阅读,论文笔记)