Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Dec 22, 2025

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline

Published on: January 30, 2019

FormNet: Formatted Learning for Image Restoration.

Jianbo Jiao, Wei-Chih Tu, Ding Liu

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society

|May 5, 2020

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

AvatarVTON: 4D Virtual Try-On for Animatable Avatars.

IEEE transactions on visualization and computer graphics·2026

Same author

Independent and joint effects of ambient temperature and relative humidity on fetal distress: effect modification by air pollutants.

International journal of biometeorology·2026

Same author

How Far Are We from Generating Missing Modalities with Foundation Models?

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Permeability prediction of 3D complex fracture networks in deep underground based on fractal analysis of trace maps.

Scientific reports·2026

Same author

Therapeutic leukapheresis with severe pertussis: evaluating efficacy, feasibility, and safety.

BMC pediatrics·2026

Same author

GranSSG: Correlating Volumetric Granularities for 3D Semantic Scene Graph Prediction.

IEEE transactions on visualization and computer graphics·2026

Same journal

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

GoP-based Quality Enhancement on Video Compression.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

Align then Tensorize: Multi-Level Consistent Anchor Graph Learning for Scalable Multi-View Clustering.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

Beyond Fidelity: Diverse Image Synthesis via Retrieval-Augmented Diffusion.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

See all related articles

This study introduces a novel deep convolutional neural network (CNN) for image restoration, enhancing structured information learning to improve performance and convergence speed. The method achieves superior quantitative and qualitative results on public datasets.

Area of Science:

Computer Vision
Deep Learning
Image Processing

Background:

Traditional deep learning methods for image restoration face challenges like gradient issues.
Existing approaches may not fully leverage the shared information between corrupted and clean images.

Purpose of the Study:

To develop an improved deep CNN for image restoration by learning formatted information.
To address gradient problems and enhance performance in image restoration tasks.

Main Methods:

Proposing a deep CNN that learns structured details and recovers latent clean images simultaneously.
Incorporating a residual formatting layer and an adversarial block to structure information.
Utilizing a cross-level loss network for both pixel-level accuracy and semantic quality.

More Related Videos

Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging

Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging

Published on: July 11, 2025

Related Experiment Videos

Last Updated: Dec 22, 2025

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline

Published on: January 30, 2019

Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging

Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging

Published on: July 11, 2025

Main Results:

The proposed method demonstrates faster network convergence.
The approach significantly boosts overall performance in image restoration.
Evaluations show favorable quantitative and qualitative results compared to existing methods.

Conclusions:

The novel deep CNN effectively tackles image restoration by learning formatted information.
The integration of residual formatting and adversarial blocks enhances network efficiency and performance.
The cross-level loss network ensures comprehensive image quality improvements.