Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Effects of concurrent transcranial direct current stimulation and robotic lower-limb training on motor recovery after stroke: a randomized sham-controlled trial.

Scientific reports·2026
Same author

Status of tetanus prevention following injuries: a descriptive study from Sandun Town, Hangzhou.

Frontiers in public health·2026
Same author

Plant Virus and Vector Insect Regulate the Dual Phosphorylation of ATG1-ATG13 To Maintain a Moderate Autophagy for Viral Persistent Propagation.

Journal of agricultural and food chemistry·2026
Same author

Reassortment risk of clade 2.3.4.4b H5N1 with avian H9N2 and human H3N2 and H1N1 influenza A viruses.

Science bulletin·2025
Same author

Selective Classification Under Distribution Shifts.

Transactions on machine learning research·2025
Same author

Dual analysis of postural control in middle-aged and elderly patients with cervicogenic dizziness: Dynamic and static balance perspectives.

Frontiers in bioengineering and biotechnology·2025
Same journal

Scaling Up Bayesian Neural Networks with Neural Networks.

Transactions on machine learning research·2026
Same journal

Tumor-anchored deep feature random forests for out-of-distribution detection in lung cancer segmentation.

Transactions on machine learning research·2026
Same journal

Still Competitive: Revisiting Recurrent Models for Irregular Time Series Prediction.

Transactions on machine learning research·2026
Same journal

Multi-Modal Foundation Models for Computational Pathology: A Survey.

Transactions on machine learning research·2026
Same journal

Sparse-Input Neural Network using Group Concave Regularization.

Transactions on machine learning research·2026
Same journal

Bayesian Neighborhood Adaptation for Graph Neural Networks.

Transactions on machine learning research·2026
See all related articles

Related Experiment Video

Updated: Mar 13, 2026

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
07:05

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures

Published on: February 15, 2022

3.0K

A Baseline Method for Removing Invisible Image Watermarks using Deep Image Prior.

Hengyue Liang1, Taihui Li2, Ju Sun2

  • 1Electrical and Computer Engineering, University of Minnesota, Twin Cities.

Transactions on Machine Learning Research
|March 12, 2026
PubMed
Summary
This summary is machine-generated.

This study introduces a simple method using Deep Image Prior (DIP) to remove invisible image watermarks from single images. This technique aids in detecting AI-generated content and benchmarking watermark robustness.

More Related Videos

Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging
07:15

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

Published on: July 11, 2025

3.6K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

1.2K

Related Experiment Videos

Last Updated: Mar 13, 2026

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
07:05

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures

Published on: February 15, 2022

3.0K
Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging
07:15

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

Published on: July 11, 2025

3.6K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

1.2K

Area of Science:

  • Computer Vision
  • Digital Image Processing
  • Artificial Intelligence Security

Background:

  • AI-generated content detection is crucial for copyright protection and preventing misinformation.
  • Invisible image watermarks are a promising but vulnerable technique for content authentication.
  • Existing watermark removal methods often require extensive datasets or knowledge of the watermarking system.

Purpose of the Study:

  • To develop a black-box method for removing invisible image watermarks.
  • To evaluate the effectiveness of Deep Image Prior (DIP) for watermark evasion.
  • To propose DIP as a baseline for benchmarking watermark robustness and discuss implications for visible watermarks.

Main Methods:

  • A black-box approach using Deep Image Prior (DIP) for watermark removal.
  • Regressing a single watermarked image through DIP's intermediate steps.
  • Benchmarking against training-based visible watermarks to assess evasion limitations.

Main Results:

  • Successfully removed invisible watermarks from single images using DIP.
  • Achieved high image quality preservation during the watermark evasion process.
  • Demonstrated the limitations of black-box methods against training-based visible watermarks.

Conclusions:

  • DIP offers a practical and effective method for evading invisible image watermarks.
  • DIP should be integrated as a baseline for evaluating watermark robustness.
  • Visible watermarks show potential for combating misinformation due to current evasion method limitations.