hellloxiaotian / Brdnet
Image denoising using deep CNN with batch renormalization(Neural Networks,2020)
Stars: ✭ 141
Projects that are alternatives of or similar to Brdnet
DLSS
Deep Learning Super Sampling with Deep Convolutional Generative Adversarial Networks.
Stars: ✭ 88 (-37.59%)
Mutual labels: cnn, cnn-keras
Image classifier
CNN image classifier implemented in Keras Notebook 🖼️.
Stars: ✭ 139 (-1.42%)
Mutual labels: cnn, cnn-keras
Screenshot To Code
A neural network that transforms a design mock-up into a static website.
Stars: ✭ 13,561 (+9517.73%)
Mutual labels: cnn, cnn-keras
Vincent Ai Artist
Style transfer using deep convolutional neural nets
Stars: ✭ 176 (+24.82%)
Mutual labels: cnn, cnn-keras
Hyperopt Keras Cnn Cifar 100
Auto-optimizing a neural net (and its architecture) on the CIFAR-100 dataset. Could be easily transferred to another dataset or another classification task.
Stars: ✭ 95 (-32.62%)
Mutual labels: cnn, cnn-keras
Personality Detection
Implementation of a hierarchical CNN based model to detect Big Five personality traits
Stars: ✭ 338 (+139.72%)
Mutual labels: cnn, cnn-keras
Resnetcam Keras
Keras implementation of a ResNet-CAM model
Stars: ✭ 269 (+90.78%)
Mutual labels: cnn, cnn-keras
Object Localization
Object localization in images using simple CNNs and Keras
Stars: ✭ 130 (-7.8%)
Mutual labels: cnn, cnn-keras
Kaggle Web Traffic Time Series Forecasting
Solution to Kaggle - Web Traffic Time Series Forecasting
Stars: ✭ 29 (-79.43%)
Mutual labels: cnn, cnn-keras
Benchmarking Keras Pytorch
🔥 Reproducibly benchmarking Keras and PyTorch models
Stars: ✭ 346 (+145.39%)
Mutual labels: cnn, cnn-keras
Keras transfer cifar10
Object classification with CIFAR-10 using transfer learning
Stars: ✭ 120 (-14.89%)
Mutual labels: cnn, cnn-keras
I3d finetune
TensorFlow code for finetuning I3D model on UCF101.
Stars: ✭ 128 (-9.22%)
Mutual labels: cnn
Adnet
Attention-guided CNN for image denoising(Neural Networks,2020)
Stars: ✭ 135 (-4.26%)
Mutual labels: cnn
Image Caption Generator
A neural network to generate captions for an image using CNN and RNN with BEAM Search.
Stars: ✭ 126 (-10.64%)
Mutual labels: cnn-keras
Keras Vgg16 Places365
Keras code and weights files for the VGG16-places365 and VGG16-hybrid1365 CNNs for scene classification
Stars: ✭ 138 (-2.13%)
Mutual labels: cnn
Easyocr
Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin, Chinese, Arabic, Devanagari, Cyrillic and etc.
Stars: ✭ 13,379 (+9388.65%)
Mutual labels: cnn
Pytorch convlstm
convolutional lstm implementation in pytorch
Stars: ✭ 126 (-10.64%)
Mutual labels: cnn
https://www.sciencedirect.com/science/article/pii/S0893608019302394) and it is implemented by Keras.
Image denoising using deep CNN with batch renormalization(BRDNet)by Chunwei Tian, Yong Xu and Wangmeng Zuo is publised in Neural Networks, 2020. (https://mp.weixin.qq.com/s/Jk6PlRBYorLI5FSa5xxOkw and https://mp.weixin.qq.com/s/dSCRx-6QW9bFDYQkDBGdLw.
This paper is pushed on home page of the Nueral Networks and BRDNet is collected by iHub,which is Chinese largest artificial intelligence platform. Also, it becomes a ESI highly cited paper. Additionally, it is reported by wechat public accounts atThis paper is the first paper via using enlaring the network width for addressing image denoising. Also, it is the first paper via deep network to address real noisy images of CC.
Absract
http://www.yongxu.org/lunwen.html.
Deep convolutional neural networks (CNNs) have attracted great attention in the field of image denoising. However, there are two drawbacks: (1) It is very difficult to train a deeper CNN for denoising tasks, and (2) most of deeper CNNs suffer from performance saturation. In this paper, we report the design of a novel network called a batch-renormalization denoising network (BRDNet). Specifically, we combine two networks to increase the width of the network, and thus obtain more features. Because batch renormalization is fused into BRDNet, we can address the internal covariate shift and small mini-batchproblems. Residual learning is also adopted in a holistic way to facilitate network training. Dilated convolutions are exploited to extract more information for denoising tasks. Extensive experimental results show that BRDNet outperforms state-of-the-art image denoising methods. The code of BRDNet is accessible atRequirements (Keras)
Tensorflow 1.3.0
Keras 2.0
Numpy
Opencv 2
Python 2.7
Cuda 8.0
Cudnn 7.5
Ubuntu 14.04
Commands
Training for gray noisy images
python mainimprovement.py
Training for color noisy images
python mainimprovement.py
Test for gray noisy images---test gray noisy image with noise level of 25
python mainimprovement.py --only_test True --pretrain 25/model_50.h5
Test for color noisy images---test color noisy image with noise level of 25
python mainimprovementcolor.py --only_test True --pretrain 25/model_50.h5
Training datasets
https://pan.baidu.com/s/13jDZfayiM-KxLAOVWvFlUA
The training dataset of the gray noisy images is downloaded athttps://pan.baidu.com/s/1cx3ymsWLIT-YIiJRBza24Q
The training dataset of the color noisy images is downloaded atNetwork architecture
Resluts
Gaussian gray noisy image denoising
Average PSNR (dB) results of different methods on BSD68 dataset with noise levels of 15, 25 and 50.
PSNR (dB) results for different methods on 12 widely used images with noise levels of 15, 25 and 50.
Visual results for gray noisy images
Denoising results of one image from the BSD68 dataset with noise level 25 using for different methods: (a) original image, (b) noisy image /20.30 dB, (c) WNNM/29.75 dB, (d) E-PLL/29.59 dB, (e) TNRD/29.76 dB, (f) DnCNN/30.16 dB, (g) BM3D/29.53 dB, (h) IRCNN/30.07 dB, and(i) BRDNet/30.27 dB.
Denoising results of image “monar” from Set12 with noise level 50 using different methods: (a) original image, (b) noisy image/14.71 dB, (c) WNNM/26.32 dB, (d) EPLL/25.94dB, (e) TNRD/26.31 dB, (f) DnCNN/26.78 dB, (g) BM3D/25.82 dB, (h) IRCNN/26.61 dB, and(i) BRDNet/26.97 dB.
Gaussian color noisy image Denoising
Average PSNR (dB) results of different methods on the CBSD68, Kodak24, and McMaster datasets with noise levels of 15, 25, 35, 50, and 75.
Visual results for color noisy images
Denoising results for one color image from the McMaster dataset with noise level 35: (a) original image/ σ = 35, (b) noisy image/18.62 dB, (c) CBM3D/31.04 dB, (d) FFDNet/31.94dB, and (e) BRDNet/32.25 dB.
Denoising results for one color image from the Kodak24 dataset with noise level 60:(a) original image/ σ = 60, (b) noisy image/13.45 dB, (c) CBM3D/31.00 dB, (d) FFDNet/31.49 dB, and (e) BRDNet/31.85 dB.
Real noisy image denoising
PSNR (dB) results for different methods on real noisy images.
Complexity and complexity of different methods for image denoising
Complexity analysis of BRDNet, DnCNN and two DnCNNs.
Running time of different methods on an image different size
Running time for different methods in denoising images of sizes 256 × 256, 512 × 512, and 1024 × 1024.
If you want to cite this paper, please refer to the following format
1. Tian C, Xu Y, Zuo W. Image denoising using deep CNN with batch renormalization[J]. Neural Networks, 2020, 121: 461-473.
@article{tian2020image,
2.title={Image denoising using deep CNN with batch renormalization},
author={Tian, Chunwei and Xu, Yong and Zuo, Wangmeng},
journal={Neural Networks},
volume={121},
pages={461--473},
year={2020},
publisher={Elsevier}
}
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].