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

Counterfactual Thinking01:19

Counterfactual Thinking

18
Counterfactual thinking is a cognitive process wherein individuals mentally reconstruct alternative versions of past events, often beginning with “what if” or “if only.” This reflective mechanism plays a significant role in shaping emotional experiences and guiding future behavior. Though typically triggered by unfavorable or unexpected outcomes, counterfactual thinking can also emerge in mundane, everyday decisions and experiences, revealing its deep entrenchment in...
18
Aliasing01:18

Aliasing

267
Accurate signal sampling and reconstruction are crucial in various signal-processing applications. A time-domain signal's spectrum can be revealed using its Fourier transform. When this signal is sampled at a specific frequency, it results in multiple scaled replicas of the original spectrum in the frequency domain. The spacing of these replicas is determined by the sampling frequency.
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
267
Hindsight Biases01:12

Hindsight Biases

4.0K
Hindsight bias leads you to believe that the event you just experienced was predictable, even though it really wasn’t. In other words, you knew all along that things would turn out the way they did. Can you relate this to the phrase "Hindsight is 20/20" now? 
4.0K
Vision01:24

Vision

55.5K
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
55.5K
Reconstruction of Signal using Interpolation01:10

Reconstruction of Signal using Interpolation

387
Signal processing techniques are essential for accurately converting continuous signals to digital formats and vice versa. When a continuous signal is sampled with a period T, the resulting sampled signal exhibits replicas of the original spectrum in the frequency domain, spaced at intervals equal to the sampling frequency. To handle this sampled signal, a zero-order hold method can be applied, which creates a piecewise constant signal by retaining each sample's value until the next...
387
Photoreceptors and Visual Pathways01:22

Photoreceptors and Visual Pathways

6.6K
At the molecular level, visual signals trigger transformations in photopigment molecules, resulting in changes in the photoreceptor cell's membrane potential. The photon's energy level is denoted by its wavelength, with each specific wavelength of visible light associated with a distinct color. The spectral range of visible light, classified as electromagnetic radiation, spans from 380 to 720 nm. Electromagnetic radiation wavelengths exceeding 720 nm fall under the infrared category,...
6.6K

You might also read

Related Articles

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

Sort by
Same author

Ash dieback and hydrology affect tree growth patterns under climate change in European floodplain forests.

Scientific reports·2025
Same author

Exploration of using a wall-climbing robot system for indoor inspection in occupied buildings.

Scientific reports·2024
Same author

Molecular phylogenetics, PCA, and MFA recover a new species of Cyrtodactylus (Squamata: Gekkonidae) from an isolated sandstone massif in northwestern Cambodia.

Zootaxa·2021
Same author

Analyzing the Influence of Hyper-parameters and Regularizers of Topic Modeling in Terms of Renyi Entropy.

Entropy (Basel, Switzerland)·2020
Same author

Counting stripes: revision of the <i>Lipinia vittigera</i> complex (Reptilia, Squamata, Scincidae) with description of two new species from Indochina.

Zoological research·2019
Same author

Integrative Taxonomy of Southeast Asian Snail-Eating Turtles (Geoemydidae: Malayemys) Reveals a New Species and Mitochondrial Introgression.

PloS one·2016

Related Experiment Video

Updated: Sep 29, 2025

Visualizing Visual Adaptation
04:43

Visualizing Visual Adaptation

Published on: April 24, 2017

9.1K

COIN: Counterfactual Image Generation for Visual Question Answering Interpretation.

Zeyd Boukhers1, Timo Hartmann1, Jan Jürjens1,2

  • 1Faculty of Computer Science, University of Koblenz-Landau, 56070 Koblenz, Germany.

Sensors (Basel, Switzerland)
|March 26, 2022
PubMed
Summary

This study introduces a new method for understanding Visual Question Answering (VQA) models by creating slightly altered, realistic images that change the VQA model's answer, aiding in interpreting complex question behaviors.

Keywords:
GANML interpretabilityUXEVQA

More Related Videos

Using Eye Movements Recorded in the Visual World Paradigm to Explore the Online Processing of Spoken Language
09:27

Using Eye Movements Recorded in the Visual World Paradigm to Explore the Online Processing of Spoken Language

Published on: October 13, 2018

10.2K
Creating Virtual-hand and Virtual-face Illusions to Investigate Self-representation
06:53

Creating Virtual-hand and Virtual-face Illusions to Investigate Self-representation

Published on: March 1, 2017

13.4K

Related Experiment Videos

Last Updated: Sep 29, 2025

Visualizing Visual Adaptation
04:43

Visualizing Visual Adaptation

Published on: April 24, 2017

9.1K
Using Eye Movements Recorded in the Visual World Paradigm to Explore the Online Processing of Spoken Language
09:27

Using Eye Movements Recorded in the Visual World Paradigm to Explore the Online Processing of Spoken Language

Published on: October 13, 2018

10.2K
Creating Virtual-hand and Virtual-face Illusions to Investigate Self-representation
06:53

Creating Virtual-hand and Virtual-face Illusions to Investigate Self-representation

Published on: March 1, 2017

13.4K

Area of Science:

  • Artificial Intelligence
  • Computer Vision
  • Natural Language Processing

Background:

  • Visual Question Answering (VQA) systems are advancing but struggle with complex questions.
  • Understanding VQA model behavior is crucial before relying on their outputs.

Purpose of the Study:

  • To develop an interpretability approach for VQA models.
  • To generate realistic counterfactual images that alter VQA model predictions.

Main Methods:

  • Generating minimal, realistic image modifications to elicit different VQA answers.
  • Conducting a user study to evaluate the interpretability approach due to the lack of quantitative metrics.

Main Results:

  • The approach successfully generates counterfactual images that change VQA model predictions.
  • User study results provide insights into VQA model behavior and the effectiveness of the interpretability method.

Conclusions:

  • The proposed counterfactual image generation offers a viable method for VQA model interpretability.
  • The approach enhances understanding of how VQA models handle complex visual questions.