Pyramid Real Image Denoising Network

Zhao, Yiyun; Jiang, Zhuqing; Men, Aidong; Ju, Guodong

doi:10.1109/vcip47243.2019.8965754

Cited by 35 publications

(25 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prior methods such as BM3D [9] and non-local means [6] rely on hand-engineered algorithms [10,13,45,46]. With the introduction of data-driven neural network-based methods [40,31,48,21,51,32], datasets with paired noisy and noise-free images became sought after [1,8,2]. However, it is well known and demonstrated in public benchmarks that performing denoising in the RAW domain yields superior results than denoising on the final RGB images [33].…”

Section: Related Workmentioning

confidence: 99%

“…Image denoising is a classical yet actively studied topic in image restoration [9,6,10,13,45,46]. Recent at-tention for the problem focuses on applying deep neural networks to RAW sensor data or to images obtained after post-processing with the device's image signal processor (ISP) [5,40,31,48,21,51,32,47]. Most of these networks are based on the U-Net [34] architecture.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Feature-Align Network with Knowledge Distillation for Efficient Denoising

Lucas¹,

Reda²,

Ranjan³

et al. 2021

Preprint

View full text Add to dashboard Cite

We propose an efficient neural network for RAW image denoising. Although neural network-based denoising has been extensively studied for image restoration, little attention has been given to efficient denoising for compute limited and power sensitive devices, such as smartphones and smartwatches. In this paper, we present a novel architecture and a suite of training techniques for high quality denoising in mobile devices. Our work is distinguished by three main contributions. (1) Feature-Align layer that modulates the activations of an encoder-decoder architecture with the input noisy images. The auto modulation layer enforces attention to spatially varying noise that tend to be "washed away" by successive application of convolutions and nonlinearity. (2) A novel Feature Matching Loss that allows knowledge distillation from large denoising networks in the form of a perceptual content loss. (3) Empirical analysis of our efficient model trained to specialize on different noise subranges. This opens additional avenue for model size reduction by sacrificing memory for compute. Extensive experimental validation shows that our efficient model produces high quality denoising results that compete with stateof-the-art large networks, while using significantly fewer parameters and MACs. On the Darmstadt Noise Dataset benchmark, we achieve a PSNR of 48.28dB, while using 263× fewer MACs, and 17.6× fewer parameters than the state-of-the-art network, which achieves 49.12dB. * Equal contribution. † Affiliated with Facebook at the time of this work.(a) Noisy Input, PSNR = 18.76 (b) Ground Truth (c) CycleISP [47], PSNR = 40.44, 702.73 GMACs/MP (d) Our Model, PSNR = 36.33, 2.67 GMACs/MP

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Feature-Align Network with Knowledge Distillation for Efficient Denoising

Lucas¹,

Reda²,

Ranjan³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Raw Image Denoising with Unified Bayer Pattern and Multiscale Strategies The team utilized a pyramid denoising network [52] and Bayer pattern unification techniques [25] where all input noisy rawRGB images are unified to RGGB bayer pattern according to the metadata information. Then inputs are passed into Squeeze-and-Excitation blocks [18] to extract features and assign weights to different channels.…”

Section: Boe-iot-aibdmentioning

confidence: 99%

NTIRE 2019 Challenge on Real Image Denoising: Methods and Results

Abdelhamed

Timofte²,

Brown³

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Self Cite

View full text Add to dashboard Cite

This paper reviews the NTIRE 2020 challenge on real image denoising with focus on the newly introduced dataset, the proposed methods and their results. The challenge is a new version of the previous NTIRE 2019 challenge on real image denoising that was based on the SIDD benchmark. This challenge is based on a newly collected validation and testing image datasets, and hence, named SIDD+. This challenge has two tracks for quantitatively evaluating image denoising performance in (1) the Bayer-pattern rawRGB and (2) the standard RGB (sRGB) color spaces. Each track ∼250 registered participants. A total of 22 teams, proposing 24 methods, competed in the final phase of the challenge. The proposed methods by the participating teams represent the current state-of-the-art performance in image denoising targeting real noisy images. The newly collected SIDD+ datasets are publicly available at: https://bit.ly/siddplus_data. A. Abdelhamed (kamel@eecs.yorku.ca, York University), M. Afifi, R. Timofte, and M.S. Brown are the NTIRE 2020 challenge organizers, while the other authors participated in the challenge. Appendix A contains the authors' teams and affiliations. NTIRE webpage: arXiv:2005.04117v1 [cs.CV] 8 May 2020

show abstract

“…Although deep-learning-based methods have been applied to the segmentation of breast tubules [8][9][10][11][12], most of them have not taken into account the reuse and fusion of multilevel features to better describe the tubule ROI regions of complex structures, diverse shapes, and different sizes for a more accurate segmentation performance. In this paper, we propose a novel semantic segmentation model named DKS-DoubleU-Net by integrating a DenseNet [23] module and a Kernel Selecting Module (KSM) [24] into the DoubleU-Net model for accurate segmentation of breast tubules in H&E images. The adopted DenseNet encoder utilizes dense features between layers to discover intrinsic characteristics and strengthen the feature propagation of breast tubules.…”

Section: Introductionmentioning

confidence: 99%

Segmentation of Breast Tubules in H&E Images Based on a DKS-DoubleU-Net Model

Chen

Zhou

Chen

et al. 2022

BioMed Research International

View full text Add to dashboard Cite

The formation of breast tubules plays an important role in the pathological grading of breast cancer. Breast tubules surrounded by a large number of epithelial cells are located in the subcutaneous tissue of the chest. The shapes of breast tubules are various, including tubular, round, and oval, which makes the process of breast tubule segmentation a difficult task. Deep learning technology, capable of learning complex data structures via efficient representation, could help pathologists accurately detect breast tubules in hematoxylin and eosin (H&E) stained images. In this paper, we propose a deep learning model named DKS-DoubleU-Net to accurately segment breast tubules with complex appearances in H&E images. The proposed DKS-DoubleU-Net model suggests using a DenseNet module as the encoder of the second subnetwork of DoubleU-Net, which utilizes dense features between layers and strengthens the propagation of features extracted in all previous layers, in order to better discover the intrinsic characteristics of breast tubules with complex structures and diverse shapes. Moreover, a feature fusing module called Kernel Selecting Module (KSM) is inserted before each output layer of the two U-Net branches of the DoubleU-Net, to implement a multiscale feature fusion via a self-adaptive kernel selecting for the sake of accurate segmentation of breast tubules in different sizes. The experiments on the public BRACS dataset and a private clinical dataset have shown that our model achieves better segmentation performance, compared to the state-of-art models of U-Net, DoubleU-Net, ResUnet++, HRNet, and DeepLabV3+. Specifically, on the public BRACS dataset, our method produced an F1-Score of 92.98%, which outperforms the F1-Score of U-Net, DoubleU-Net, and HRNet by 4.24%, 0.37%, and 1.68%, respectively, and is much better than performances of DeepLabV3+ and ResUnet++ by 7.83% and 23.84%, respectively. On the private clinic dataset, the proposed model achieved an F1-Score of 73.13%, which has shown an improvement of 10.31%, 1.89%, 4.88%, 15.47%, and 31.1% to the performances of the U-Net, DoubleU-Net, HRNet, DeepLabV3+, and ResUnet++, respectively. Superior performance could also be observed when comparing the proposed DKS-DoubleU-Net with the others using the metrics of Dice and mIou.

show abstract

Pyramid Real Image Denoising Network

Cited by 35 publications

References 18 publications

Feature-Align Network with Knowledge Distillation for Efficient Denoising

Feature-Align Network with Knowledge Distillation for Efficient Denoising

NTIRE 2019 Challenge on Real Image Denoising: Methods and Results

Segmentation of Breast Tubules in H&E Images Based on a DKS-DoubleU-Net Model

Contact Info

Product

Resources

About