When Image Denoising Meets High-Level Vision Tasks: A Deep Learning Approach

  title={When Image Denoising Meets High-Level Vision Tasks: A Deep Learning Approach},
  author={Ding Liu and Bihan Wen and Xianming Liu and Thomas S. Huang},
  booktitle={International Joint Conference on Artificial Intelligence},
Conventionally, image denoising and high-level vision tasks are handled separately in computer vision. [] Key Method Second we propose a deep neural network solution that cascades two modules for image denoising and various high-level tasks, respectively, and use the joint loss for updating only the denoising network via back-propagation. We demonstrate that on one hand, the proposed denoiser has the generality to overcome the performance degradation of different high-level vision tasks. On the other hand…

Figures and Tables from this paper

Connecting Image Denoising and High-Level Vision Tasks via Deep Learning

A convolutional neural network in which convolutions are conducted in various spatial resolutions via downsampling and upsampling operations in order to fuse and exploit contextual information on different scales and the benefit of exploiting image semantics simultaneously for image denoising and high-level vision tasks via deep learning is demonstrated.

Combining Multi-level Loss for Image Denoising

An end-to-end multi-module neural network architecture is proposed, which introduces the combination of the high-level and low-level feature in the training process, for image denoising and introduces a hybrid loss, which is designed to restore details from both pixel and feature space.

ResDNN: deep residual learning for natural image denoising

A deep convolution neural network with added benefits of residual learning for image denoising is proposed, capable of learning end-to-end mappings from noise distorted images to restored cleaner versions.

Real-World Image Denoising with Deep Boosting

A lightweight Dense Dilated Fusion Network (DDFN) is designed as an embodiment of the boosting unit, which addresses the vanishing of gradients during training due to the cascading of networks while promoting the efficiency of limited parameters.

Cross-Patch Graph Convolutional Network for Image Denoising

This paper proposes Cross-Patch Net (CPNet), which is the first deep-learning-based real image denoising method for HR (high resolution) input and achieves state-of-the-art performance on practical sRGB noisy image Denoising.

Semantic-Guided Zero-Shot Learning for Low-Light Image/Video Enhancement

  • Shen ZhengGaurav Gupta
  • Computer Science
    2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)
  • 2022
A semantic-guided zero-shot low-light enhancement network (SGZ) which is trained in the absence of paired images, unpaired datasets, and segmentation annotation is proposed and an unsupervised semantic segmentation network is introduced for preserving the semantic information during intensive enhancement.

Deep Learning on Image Denoising: An overview

Multinoise-type blind denoising using a single uniform deep convolutional neural network

Extensive experiments on synthetic noisy images validate the effectiveness of the noise-estimation and denoising subnetworks and show that MBDNet is highly competitive as compared to state-of-the-art methods in both denoised performance and model runtime.

Parallel Capsule Network: A Novel Method for Image Denoising

This work presents a novel denoising image method that is based on image classification, the result of image classification can be regarded as prior information for imageDenoising, and evaluates the model on the MNIST dataset and FASHION MNIST datasets.

Connecting Low-Level Image Processing and High-Level Vision via Deep Learning

The goal is to design deep learning mechanisms that can efficiently and effectively learn features from low-level image processing and use them to improve the performance of high-level vision tasks.



Image Denoising and Inpainting with Deep Neural Networks

A novel approach to low-level vision problems that combines sparse coding and deep networks pre-trained with denoising auto-encoder (DA) is presented and can automatically remove complex patterns like superimposed text from an image, rather than simple patterns like pixels missing at random.

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

This paper investigates the construction of feed-forward denoising convolutional neural networks (DnCNNs) to embrace the progress in very deep architecture, learning algorithm, and regularization method into image Denoising, and uses residual learning and batch normalization to speed up the training process as well as boost theDenoising performance.

Learning Deep CNN Denoiser Prior for Image Restoration

Experimental results demonstrate that the learned set of denoisers can not only achieve promising Gaussian denoising results but also can be used as prior to deliver good performance for various low-level vision applications.

Natural Image Denoising with Convolutional Networks

An approach to low-level vision is presented that combines the use of convolutional networks as an image processing architecture and an unsupervised learning procedure that synthesizes training samples from specific noise models to avoid computational difficulties in MRF approaches that arise from probabilistic learning and inference.

Robust Single Image Super-Resolution via Deep Networks With Sparse Prior

It is demonstrated that a sparse coding model particularly designed for SR can be incarnated as a neural network with the merit of end-to-end optimization over training data and significantly outperforms the existing state-of-the-art methods for various scaling factors both quantitatively and perceptually.

Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections

This paper proposes to symmetrically link convolutional and de-convolutional layers with skip-layer connections, with which the training converges much faster and attains a higher-quality local optimum, making training deep networks easier and achieving restoration performance gains consequently.

Enhance Visual Recognition Under Adverse Conditions via Deep Networks

A deep learning-based framework for improving the performance of image and video recognition models under adverse conditions, using robust adverse pre-training or its aggressive variant, and develops a transfer learning approach to cope with real-world datasets of unknown adverse conditions.

On the Relation between Color Image Denoising and Classification

A novel deep learning architecture for color (multichannel) image denoising and report on thousands of images from ImageNet dataset as well as commonly used imagery, finding that it is beneficial to incorporate in classification models.

Non-local sparse models for image restoration

Experimental results in image denoising and demosaicking tasks with synthetic and real noise show that the proposed method outperforms the state of the art, making it possible to effectively restore raw images from digital cameras at a reasonable speed and memory cost.

Very Deep Convolutional Networks for Large-Scale Image Recognition

This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.