computer-vision-5

Reflection Removal

3260 papers • 126 benchmarks • 313 datasets

Remove the spots from mirror and clear the picture

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in reflection-removal-11

Trend

Dataset

Best Model

Actions

Real20

SIR^2(Objects)

SIR^2(Postcard)

Libraries

i

Use these libraries to find reflection-removal-11 models and implementations

Datasets

CDR

SlowFlow

PolarRR

Subtasks

No subtasks available.

Most implemented papers

Single Image Reflection Removal Using Deep Encoder-Decoder Network

Jinjin Gu, Zhixiang Chi, Xiaolin Wu, X. Shu•Tue Jan 30 2018

A novel deep convolutional encoder-decoder method to remove the objectionable reflection by learning a map between image pairs with and without reflection, which significantly outperforms the other tested state-of-the-art techniques.

30

Content

SIR^2(Wild)

Nature

0

Paper Graph

Single Image Reflection Separation with Perceptual Losses

Ren Ng, Qifeng Chen, X. Zhang•Thu May 31 2018

The approach uses a fully convolutional network trained end-to-end with losses that exploit low-level and high-level image information and proposes a novel exclusion loss that enforces pixel-level layer separation.

369 0

Paper Graph

Single Image Reflection Removal Through Cascaded Refinement

Stephen Lin, Chao Li, Yixiao Yang, Kun He, J. Hopcroft•Thu Nov 14 2019

This work proposes an Iterative Boost Convolutional LSTM Network (IBCLN) that enables cascaded prediction for reflection removal and creates a dataset of real-world images with reflection and ground-truth transmission layers to mitigate the problem of insufficient data.

150 0

Paper Graph

A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothing

Jiaolong Yang, G. Hua, Baoquan Chen, Qingnan Fan, D. Wipf•Thu Aug 10 2017

A deep neural network structure that exploits edge information in addressing representative low-level vision tasks such as layer separation and image filtering by estimating edges and reconstructing images using only cascaded convolutional layers arranged such that no handcrafted or application-specific image-processing components are required.

309 0

Paper Graph

CRRN: Multi-scale Guided Concurrent Reflection Removal Network

Ling-yu Duan, A. Kot, Boxin Shi, Renjie Wan, A. Tan•Tue May 29 2018

This paper proposes the Concurrent Reflection Removal Network (CRRN), a network that integrates image appearance information and multi-scale gradient information with human perception inspired loss function, and is trained on a new dataset with 3250 reflection images taken under diverse real-world scenes.

123 0

Paper Graph

Seeing Deeply and Bidirectionally: A Deep Learning Approach for Single Image Reflection Removal

Javen Qinfeng Shi, Jie Yang, Lingqiao Liu, Dong Gong•Fri Sep 07 2018

This work argues that, to remove reflection truly well, it should estimate the reflection and utilize it to estimate the background image, and proposes a cascade deep neural network, which estimates both the Background image and the reflection.

176 0

Paper Graph

Fourier-Domain Optimization for Image Processing

R. Achanta, S. Süsstrunk, Majed El Helou, Frederike Dümbgen•Mon Sep 10 2018

The mathematical background and motivating why the presented setups can be transformed and solved very efficiently in the Fourier domain are explained and how to practically use these solutions is shown, by providing the corresponding implementations.

11 0

Paper Graph

Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancements

Jiaolong Yang, Kaixuan Wei, Ying Fu, Hua Huang, D. Wipf•Sun Mar 31 2019

Experimental results collectively show that the method outperforms the state-of-the-art with aligned data, and that significant improvements are possible when using additional misaligned data.

178 0

Paper Graph

Single Image Reflection Removal With Physically-Based Training Images

Yuchi Huo, S. Kim, Sung-eui Yoon•Thu Apr 25 2019

Recently, deep learning-based single image reflection separation methods have been exploited widely. To benefit the learning approach, a large number of training image pairs (i.e., with and without reflections) were synthesized in various ways, yet they are away from a physically-based direction. In this paper, physically based rendering is used for faithfully synthesizing the required training images, and a corresponding network structure and loss term are proposed. We utilize existing RGBD/RGB images to estimate meshes, then physically simulate the light transportation between meshes, glass, and lens with path tracing to synthesize training data, which successfully reproduce the spatially variant anisotropic visual effect of glass reflection. For guiding the separation better, we additionally consider a module, backtrack network (BT-net) for backtracking the reflections, which removes complicated ghosting, attenuation, blurred and defocused effect of glass/lens. This enables obtaining a priori information before having the distortion. The proposed method considering additional a priori information with physically simulated training data is validated with various real reflection images and shows visually pleasant and numerical advantages compared with state-of-the-art techniques.

90 0

Paper Graph

Semantic Guided Single Image Reflection Removal

Shaodi You, Feng Lu, Yunfei Liu, Yu Li•Fri Jul 26 2019

A multi-task end-to-end deep learning method with a semantic guidance component, to solve reflection removal and semantic segmentation jointly and shows significant performance gain when using high-level object-oriented information.

20 0

Paper Graph

Adding a benchmark result helps the community track progress.