Awesome
<p align="center"> <h1 align="center">Awesome-Single-Image-Reflection-Removal(SIRR) </h1> <p align="center">A collection of Single image reflection removal. <br /> <br /> <a href="https://github.com/Liar-zzy/Awesome-Single-Image-Reflection-Removal/issues/new">Suggest new item</a> <br /> <a href="https://github.com/Liar-zzy/Awesome-Single-Image-Reflection-Removal/issues/new">Report Bug</a> </p> <p align="center"> <a href="https://github.com/Liar-zzy/Awesome-Single-Image-Reflection-Removal"> <img src= "https://img.shields.io/github/stars/Liar-zzy/Awesome-Single-Image-Reflection-Removal"></a> <img style="border-radius:5px;width:120px;" src="https://badges.toozhao.com/badges/01HGGJYNZ9DFR8JC1TPZEJJ2YQ/blue.svg" alt="Count"> </p> </p> <p align="center">:star2: If Awesome-SIRR is helpful to your images or projects, please help star this repo. Thanks! :hugs:</p>This repository provides a summary of deep learning-based SIRR algorithms. All of these methods have been verified through the DBLP.
:rocket: To-Do-List
- SIRR methods classification
- 2023.12.01 --- Update SOTA SIRR methods
Table of contents
:boom: SIRR
Year | Pub | Title | Links |
---|---|---|---|
2023 | CVPR | :fire: :fire: :fire: Robust Single Image Reflection Removal Against Adversarial Attacks | [paper]<br />[code] |
2022 | WACV | Improving Single-Image Defocus Deblurring: How Dual-Pixel Images Help Through Multi-Task Learning | [paper]<br />[code] |
2022 | PAMI | Benchmarking Single-Image Reflection Removal Algorithms | [paper]<br />[website] |
2021 | WACV | Single image reflection removal with edge guidance | [paper]<br/>[code] |
2021 | ICCV | Location-aware single image reflection removal | [paper]<br />[code] |
2021 | NIPS | Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separation | [paper]<br />[code] |
2021 | CVPR | Single Image Reflection Removal with Absorption Effect | [paper]<br />[code] |
2021 | CVPR | Robust Reflection Removal with Reflection-free Flash-only Cues | [paper]<br />[code] |
2021 | ICCV | V-DESIRR: Very Fast Deep Embedded Single Image Reflection Removal | [paper]<br />[code]✖️ |
2021 | TIP | Deep-Masking Generative Network: A Unified Framework for Background Restoration from Superimposed Images | [paper]<br />[code] |
2020 | CVPR | Single Image Reflection Removal Through Cascaded Refinement | [paper]<br />[code] |
2020 | CVPR | Single Image Reflection Removal with Physically-Based Training Images | [paper]<br />[code] |
2020 | CVPR | Deep Adversarial Decomposition: A Unified Framework for Separating Superimposed Images | [paper]<br />[code] |
2019 | CVPR | Single image reflection removal exploiting misaligned training data and network enhancements | [paper]<br />[code] |
2019 | CVPR | Single Image Reflection Removal Beyond Linearity | [paper]<br />[code] |
2019 | PAMI | CoRRN: Cooperative Reflection Removal Network | [paper]<br />[code] |
2018 | ECCV | Seeing deeply and bidirectionally: A deep learning approach for single image reflection removal | [paper]<br />[code] |
2018 | CVPR | Single image reflection separation with perceptual losses | [paper]<br />[code] |
2018 | CVPR | CRRN: Multi-Scale Guided Concurrent Reflection Removal Network | [paper]<br />[code](not official) |
2017 | ICCV | Benchmarking Single-Image Reflection Removal Algorithms | [paper]<br />[website] |
2017 | ICCV | A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothing | [paper]<br />[code] |
2014 | CVPR | Single image layer separation using relative smoothness | [paper]<br />[code] |
:hugs: Image Quality Assessment
Method | Type | Code/Ref |
---|---|---|
PSNR (Peak Signal-to-Noise Ratio) | Full-Reference | [code] |
SSIM (Structural Similarity Index Measurement) | Full-Reference | [code] |
LPIPS (Learned Perceptual Image Patch Similarity) | Full-Reference | [code] |
:+1: Benchmark Datasets
Recommended Datasets
Dataset | Usage | Image num |
---|---|---|
SIR<sup>2</sup> | Test | 500 |
Real20 | Training & Test | 89 & 20 |
Nature | Test | 20 |
CDR | Test | 1063 |
:sparkles: Acknowledgement
This work is inspired by awesome-reflection-removal and Awesome-Reflection-Removal.