Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections

Abstract

Image restoration, including image denoising, super resolution, inpainting, and so on, is a well-studied problem in computer vision and image processing, as well as a test bed for low-level image modeling algorithms. In this work, we propose a very deep fully convolutional auto-encoder network for image restoration, which is a encoding-decoding framework with symmetric convolutional-deconvolutional layers. In other words, the network is composed of multiple layers of convolution and de-convolution operators, learning end-to-end mappings from corrupted images to the original ones. The convolutional layers capture the abstraction of image contents while eliminating corruptions. Deconvolutional layers have the capability to upsample the feature maps and recover the image details. To deal with the problem that deeper networks tend to be more difficult to train, we propose to symmetrically link convolutional and deconvolutional layers with skip-layer connections, with which the training converges much faster and attains better results.

arXiv comment17 pages. A journal extension of the version at arXiv:1603.09056

Cited by in corpus (61)

• MultiResUNet : Rethinking the U-Net Architecture for Multimodal Biomedical Image Segmentation
• Hyperspectral Image Denoising Employing a Spatial-Spectral Deep Residual Convolutional Neural Network
• An Iterative BP-CNN Architecture for Channel Decoding
• Learning Implicit Brain MRI Manifolds with Deep Learning
• Deep Chain HDRI: Reconstructing a High Dynamic Range Image from a Single Low Dynamic Range Image
• Implicit Dual-domain Convolutional Network for Robust Color Image Compression Artifact Reduction
• Motion Corrected Multishot MRI Reconstruction Using Generative Networks with Sensitivity Encoding
• Bridging the Gap Between Computational Photography and Visual Recognition
• MTRNet++: One-stage Mask-based Scene Text Eraser
• Improving Electron Micrograph Signal-to-Noise with an Atrous Convolutional Encoder-Decoder
• $\text{FastSVD-ML-ROM}$: A Reduced-Order Modeling Framework based on Machine Learning for Real-Time Applications
• Just Noticeable Difference for Deep Machine Vision
• High-Quality Self-Supervised Deep Image Denoising
• Towards CT-quality Ultrasound Imaging using Deep Learning
• Convolutional Neural Pyramid for Image Processing
• Single Image Reflection Removal Using Deep Encoder-Decoder Network
• S-Net: A Scalable Convolutional Neural Network for JPEG Compression Artifact Reduction
• Optimal Physical Preprocessing for Example-Based Super-Resolution
• Deep neural networks-based denoising models for CT imaging and their efficacy
• Wide Inference Network for Image Denoising via Learning Pixel-distribution Prior
• A Provably Convergent Scheme for Compressive Sensing under Random Generative Priors
• Learning Pixel-Distribution Prior with Wider Convolution for Image Denoising
• Disentangling Factors of Variation by Mixing Them
• Painting baryons onto N-body simulations of galaxy clusters with image-to-image deep learning
• First image then video: A two-stage network for spatiotemporal video denoising
• Visual Time Series Forecasting: An Image-driven Approach
• DSAL-GAN: Denoising based Saliency Prediction with Generative Adversarial Networks
• TV-GAN: Generative Adversarial Network Based Thermal to Visible Face Recognition
• Noise reduction on single-shot images using an autoencoder
• Denoising Arterial Spin Labeling Cerebral Blood Flow Images Using Deep Learning
• View Extrapolation of Human Body from a Single Image
• DriveGuard: Robustification of Automated Driving Systems with Deep Spatio-Temporal Convolutional Autoencoder
• Learning Deep Representations Using Convolutional Auto-encoders with Symmetric Skip Connections
• ChaLearn Looking at People: Inpainting and Denoising challenges
• Deep End-to-end Fingerprint Denoising and Inpainting
• A Possible Converter to Denoise the Images of Exoplanet Candidates through Machine Learning Techniques
• Deep Learning with Energy-efficient Binary Gradient Cameras
• Long-Term Image Boundary Prediction
• Walking the Tightrope: An Investigation of the Convolutional Autoencoder Bottleneck
• Explanation based Handwriting Verification
• Deep-seismic-prior-based reconstruction of seismic data using convolutional neural networks
• A lightweight convolutional neural network for image denoising with fine details preservation capability
• Image Restoration Using Deep Regulated Convolutional Networks
• DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification
• KRNET: Image Denoising with Kernel Regulation Network
• Ultra High Fidelity Image Compression with ${\ell }_{\infty }$-constrained Encoding and Deep Decoding
• Robust Segmentation of Optic Disc and Cup from Fundus Images Using Deep Neural Networks
• Coupled-Projection Residual Network for MRI Super-Resolution
• Assessing the performance of CT image denoisers using Laguerre-Gauss Channelized Hotelling Observer for lesion detection
• Weak Texture Information Map Guided Image Super-resolution with Deep Residual Networks
• Machine Vision for Improved Human-Robot Cooperation in Adverse Underwater Conditions
• High-Resolution Pelvic MRI Reconstruction Using a Generative Adversarial Network with Attention and Cyclic Loss
• DeepA: A Deep Neural Analyzer For Speech And Singing Vocoding
• Modeling Graph Node Correlations with Neighbor Mixture Models
• Dynamic Low-light Imaging with Quanta Image Sensors
• SDCNet: Smoothed Dense-Convolution Network for Restoring Low-Dose Cerebral CT Perfusion
• Singular points detection with semantic segmentation networks
• DAVID: Dual-Attentional Video Deblurring
• Domain Translation with Conditional GANs: from Depth to RGB Face-to-Face
• Deep Learned Optical Multiplexing for Multi-Focal Plane Microscopy
• Lattice Fusion Networks for Image Denoising