AIM 2019 Challenge on Image Demoireing: Dataset and Study

Yuan, Shanxin; Timofte, Radu; Slabaugh, Greg; Leonardis, Aleš

doi:10.1109/iccvw.2019.00437

Cited by 30 publications

(33 citation statements)

References 15 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our 3DNet is implemented in PyTorch and runs on one Nvidia RTX2080Ti GPU for 36 hours. In our network, the patch size is set to 256 and Adam [10] We compare 3DNet with state-of-the-art methods mainly on the LCDmoiré [11] and TIP2018 [1] datasets and conduct extensive ablation studies to demonstrate the necessity of each component in the design. The LCDmoiré dataset contains 10,200 synthetic image pairs.…”

Section: Methodsmentioning

confidence: 99%

Image Demoiréing with a Dual-Domain Distilling Network

Wang

Guo

2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

Due to slight discrepancy of spatial frequency between the camera sensor array and sub-pixel layout of LCD monitor, moiré pattern artifacts appear in various shapes and colours which seriously degrade the quality of captured images. It is challenging yet practically crucial to remove moiré artifacts from a single camera-captured screen image. In this paper, we propose a dual-domain distilling network (3DNet for short) to tackle this problem in an end-to-end manner. The 3DNet consists of a dual-branch student network (a.k.a. demoiréing network), and two teacher networks. The two branches of student network exploit knowledge in both spatial-domain and frequency-domain for the sake of removing moiré artifacts, based on the observation that rich image details can be discovered in the frequency-domain while structure information can be well kept in the spatial-domain. The demoiréing process of two branches is supervised by the knowledge distilled from two teacher networks trained for reconstructing clear images in the spatial and frequency domains respectively. Comprehensive experimental results are conducted to demonstrate the efficacy of our design, and reveal its superiority over state-ofthe-art alternatives.

show abstract

Section: Methodsmentioning

confidence: 99%

Image Demoiréing with a Dual-Domain Distilling Network

Wang

Guo

2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

show abstract

“…In [39,53], the authors demonstrated that convolutional neural networks can be used for performing image demosaicing, and outperformed several conventional models in this task. Works [2,16] used CNNs for correcting the white balance of RGB images, and in [63] deep learning models were applied to synthetic LCDMoire dataset for solving image demoireing problem. In [49], the authors collected 110 RAW low-lit images with Samsung S7 phone, and used a CNN model to remove noise and brighten demosaiced RGB images obtained with a simple hand-designed ISP.…”

Section: Related Workmentioning

confidence: 99%

Replacing Mobile Camera ISP with a Single Deep Learning Model

Ignatov

Gool

Timofte

2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Self Cite

158

127

View full text Add to dashboard Cite

As the popularity of mobile photography is growing constantly, lots of efforts are being invested now into building complex hand-crafted camera ISP solutions. In this work, we demonstrate that even the most sophisticated ISP pipelines can be replaced with a single end-to-end deep learning model trained without any prior knowledge about the sensor and optics used in a particular device. For this, we present PyNET, a novel pyramidal CNN architecture designed for fine-grained image restoration that implicitly learns to perform all ISP steps such as image demosaicing, denoising, white balancing, color and contrast correction, demoireing, etc. The model is trained to convert RAW Bayer data obtained directly from mobile camera sensor into photos captured with a professional high-end DSLR camera, making the solution independent of any particular mobile ISP implementation. To validate the proposed approach on the real data, we collected a large-scale dataset consisting of 10 thousand full-resolution RAW-RGB image pairs captured in the wild with the Huawei P20 cameraphone (12.3 MP Sony Exmor IMX380 sensor) and Canon 5D Mark IV DSLR. The experiments demonstrate that the proposed solution can easily get to the level of the embedded P20's ISP pipeline that, unlike our approach, is combining the data from two (RGB + B/W) camera sensors. The dataset, pretrained models and codes used in this paper are available on the project website: https://people.ee.ethz. ch/˜ihnatova/pynet.html

show abstract

“…We use the LCDMoire dataset [20] provided by the AIM contest as the training and validation sets, without using any additional datasets. We use pytorch1.2 to build our proposed MDDM model and use NVIDIA Titan V GPU with CUDA10.0 to accelerate training.…”

Section: Dataset and Train Detailmentioning

confidence: 99%

Multi-Scale Dynamic Feature Encoding Network for Image Demoiréing

Cheng

Yang

2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

The prevalence of digital sensors, such as digital cameras and mobile phones, simplifies the acquisition of photos. Digital sensors, however, suffer from producing Moiré when photographing objects having complex textures, which deteriorates the quality of photos. Moiré spreads across various frequency bands of images and is a dynamic texture with varying colors and shapes, which pose two main challenges in demoiréing-an important task in image restoration. In this paper, towards addressing the first challenge, we design a multi-scale network to process images at different spatial resolutions, obtaining features in different frequency bands, and thus our method can jointly remove moiré in different frequency bands. Towards solving the second challenge, we propose a dynamic feature encoding module (DFE), embedded in each scale, for dynamic texture. Moiré pattern can be eliminated more effectively via DFE. Our proposed method, termed Multi-scale convolutional network with Dynamic feature encoding for image DeMoiréing (MDDM), can outperform the state of the arts in fidelity as well as perceptual on benchmarks.

show abstract

AIM 2019 Challenge on Image Demoireing: Dataset and Study

Cited by 30 publications

References 15 publications

Image Demoiréing with a Dual-Domain Distilling Network

Image Demoiréing with a Dual-Domain Distilling Network

Replacing Mobile Camera ISP with a Single Deep Learning Model

Multi-Scale Dynamic Feature Encoding Network for Image Demoiréing

Contact Info

Product

Resources

About