【图像去噪】论文精读:Linear Combinations of Patches Are Unreasonably Effective for Single-Image Denoising

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
  • I. INTRODUCTION
  • II. A PARAMETRIC VIEW OF TWO-STEP NON-LOCALMETHODS FOR SINGLE-IMAGE DENOISING
    • A. A Unified Framework for Non-Local Denoisers
    • B. Parameter Optimization
    • C. Principle of Internal Adaptation
  • III. LICHI: LINEAR AND ITERATIVE COMBINATIONS OFPATCHES FOR IMAGE DENOISING
    • A. A Novel Chaining Rule for Generalization
    • B. A Progressive Scheme for Parameter Optimization
    • C. Resolution When the True Image Is Available
    • D. Use of Multiple Cost-Efficient Pilots to Guide Estimation
    • E. Weighted Average Reprojection
  • IV. BUILDING AN INITIAL PILOT
    • A. Stein's Unbiased Risk Estimate (SURE)
    • B. Noisier2Noise
    • C. Two Additional Extreme Pilots
    • D. Comparison of the Pilots
    • E. The Crucial Role of the Aggregation Stage
  • V. EXPERIMENTAL RESULTS
    • A. Setting of Algorithm Parameters
    • B. Results on Artificially Noisy and Real-World Noisy Images
  • C. Complexity
  • VI. CONCLUSION


前言

论文题目:Linear Combinations of Patches Are Unreasonably Effective for Single-Image Denoising —— 斑块的线性组合对单幅图像去噪是不合理的有效

论文地址:Linear Combinations of Patches Are Unreasonably Effective for Single-Image Denoising

论文源码:https://github.com/sherbret/LIChI

TIP 2024

你可能感兴趣的:(深度学习,图像处理,计算机视觉,图像去噪,人工智能)