NTIRE 2019 Challenge on Real Image Denoising: Methods and Results

Abdelhamed, Abdelrahman; Timofte, Radu; Brown, Michael S.; Yu, Songhyun; Park, Bumjun; Jeong, Jechang; Jung, Seung-Won; Kim, Dong-Wook; Chung, Jae‐Ryun; Liu, Jiaming; Wang, Yuzhi; Wu, Chun-Yi; Xu, Qiong; Wang, Chuan; Cai, Shaofan; Ding, Yifan; Fan, Haoqiang; Wang, Jue; Zhang, Kai; Zuo, Wangmeng; Zhussip, Magauiya; Park, Dong Won; Soltanayev, Shakarim; Chun, Se Young; Xiong, Zhiwei; Chen, Chang; Haris, Muhammad; Akita, Kazutoshi; Yoshida, Tomoki; Shakhnarovich, Greg; Ukita, Norimichi; Zamir, Syed Waqas; Arora, Aditya; Khan, Salman; Khan, Fahad Shahbaz; Shao, Ling; Ko, Sung-Jea; Lim, Dongpan; Kim, Seung Wook; Ji, Seo-Won; Lee, Sang‐Won; Tang, Wenyi; Yi, Fan; Zhou, Yuqian; Liu, Ding; Huang, Thomas S.; Meng, Deyu; Zhang, Lei; Yong, Hongwei; Zhao, Yiyun; Tang, Pengliang; Lu, Yue; Schettini, Raimondo; Bianco, Simone; Zini, Simone; Li, Chi; Wang, Yang; Cao, Zhiguo

doi:10.1109/cvprw.2019.00273

Cited by 88 publications

(43 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, following a specified pipeline, RGB images can be accurately simulated from hyperspectral images (refer to section 2.2). The so-called physical consistency asks the question: if the reconstructed hyperspectral images are [32] Deep-imagelab 0.03010 (1) 0.01293 0.06216 (3) 0.01991 MDISL-lab ppplang 0.03075 (2) 0.01268 0.06212 (2) 0.01946 OrangeCat [68] zyz987 0.03231 (3) 0.01389 0.06200 (1)…”

Section: Physical Consistency Of Resultsmentioning

confidence: 99%

See 1 more Smart Citation

NTIRE 2018 Challenge on Spectral Reconstruction from RGB Images

Arad

Liu

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Self Cite

101

124

View full text Add to dashboard Cite

This paper reviews the second challenge on spectral reconstruction from RGB images, i.e., the recovery of wholescene hyperspectral (HS) information from a 3-channel RGB image. As in the previous challenge, two tracks were provided: (i) a "Clean" track where HS images are estimated from noise-free RGBs, the RGB images are themselves calculated numerically using the ground-truth HS images and supplied spectral sensitivity functions (ii) a "Real World" track, simulating capture by an uncalibrated and unknown camera, where the HS images are recovered from noisy JPEG-compressed RGB images. A new, larger-than-ever, natural hyperspectral image data set is presented, containing a total of 510 HS images. The Clean and Real World tracks had 103 and 78 registered participants respectively, with 14 teams competing in the final testing phase. A description of the proposed methods, alongside their challenge scores and an extensive evaluation of top performing methods is also provided. They gauge the state-of-the-art in spectral reconstruction from an RGB image. arXiv:2005.03412v1 [eess.IV] 7 May 2020

show abstract

Section: Physical Consistency Of Resultsmentioning

confidence: 99%

“…The RGB to spectra recovery challenge [9] is one of the NTIRE 2020 challenges. The other challenges are: deblurring [40], nonhomogeneous dehazing [5], perceptual extreme super-resolution [63], video quality mapping [18], real image denoising [1], real-world super-resolution [35] and demoireing [60].…”

Section: Ntire 2020 Challengementioning

confidence: 99%

NTIRE 2018 Challenge on Spectral Reconstruction from RGB Images

Arad

Liu

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Self Cite

101

124

View full text Add to dashboard Cite

show abstract

“…Apart from learning from synthesized dataset, we studied our proposed method with real-world noisy data samples also. Therefore, we trained our SAGAN with real-world noisy sampled images from Smartphone Image Denoising Dataset (SIDD) [2,3]. Also, we developed an android application to capture noisy images with real Nona-Bayer hardware.…”

Section: Experiments Setupmentioning

confidence: 99%

SAGAN: Adversarial Spatial-asymmetric Attention for Noisy Nona-Bayer Reconstruction

Sharif¹,

Biswas²

2021

Preprint

View full text Add to dashboard Cite

Nona-Bayer colour filter array (CFA) pattern is considered one of the most viable alternatives to traditional Bayer patterns. Despite the substantial advantages, such non-Bayer CFA patterns are susceptible to produce visual artefacts while reconstructing RGB images from noisy sensor data. This study addresses the challenges of learning RGB image reconstruction from noisy Nona-Bayer CFA comprehensively. We propose a novel spatial-asymmetric attention module to jointly learn bi-direction transformation and large-kernel global attention to reduce the visual artefacts. We combine our proposed module with adversarial learning to produce plausible images from Nona-Bayer CFA. The feasibility of the proposed method has been verified and compared with the state-of-the-art image reconstruction method. The experiments reveal that the proposed method can reconstruct RGB images from noisy Nona-Bayer CFA without producing any visually disturbing artefacts. Also, it can outperform the state-of-the-art image reconstruction method in both qualitative and quantitative comparison. Code available: https://github.com/sharif-apu/SAGAN_BMVC21.

show abstract

“…To better address the problem of real image denoising, current attempts can be roughly divided into the following categories: (1) realistic noise modeling (Shi Guo 2018;Brooks et al 2019;Abdelhamed, Timofte, and Brown 2019),…”

Section: Introductionmentioning

confidence: 99%

When AWGN-Based Denoiser Meets Real Noises

Zhou

Jiao

Huang

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

Discriminative learning based image denoisers have achieved promising performance on synthetic noises such as Additive White Gaussian Noise (AWGN). The synthetic noises adopted in most previous work are pixel-independent, but real noises are mostly spatially/channel-correlated and spatially/channel-variant. This domain gap yields unsatisfied performance on images with real noises if the model is only trained with AWGN. In this paper, we propose a novel approach to boost the performance of a real image denoiser which is trained only with synthetic pixel-independent noise data dominated by AWGN. First, we train a deep model that consists of a noise estimator and a denoiser with mixed AWGN and Random Value Impulse Noise (RVIN). We then investigate Pixel-shuffle Down-sampling (PD) strategy to adapt the trained model to real noises. Extensive experiments demonstrate the effectiveness and generalization of the proposed approach. Notably, our method achieves state-of-the-art performance on real sRGB images in the DND benchmark among models trained with synthetic noises. Codes are available at https://github.com/yzhouas/PD-Denoising-pytorch.

show abstract

NTIRE 2019 Challenge on Real Image Denoising: Methods and Results

Cited by 88 publications

References 64 publications

NTIRE 2018 Challenge on Spectral Reconstruction from RGB Images

NTIRE 2018 Challenge on Spectral Reconstruction from RGB Images

SAGAN: Adversarial Spatial-asymmetric Attention for Noisy Nona-Bayer Reconstruction

When AWGN-Based Denoiser Meets Real Noises

Contact Info

Product

Resources

About