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: Oct 23, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Do Neural Networks for Segmentation Understand Insideness?

Kimberly Villalobos¹, Vilim Štih², Amineh Ahmadinejad³

¹Center for Brains, Minds and Machines, MIT, Cambridge, MA 02139, U.S.A. kimvc@mit.edu.

Neural Computation

|August 19, 2021

Summary

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

Behavioral phenotypes and neuronal biomarkers in F1 mutant macaque model of SHANK3-associated autism spectrum disorders.

Neuron·2026

Same author

Visual motion and landmark position align with heading direction in the zebrafish interpeduncular nucleus.

Nature communications·2025

Same author

Configural processing as an optimized strategy for robust object recognition in neural networks.

Communications biology·2025

Same author

Strategies for integrating artificial intelligence into mammography screening programmes: a retrospective simulation analysis.

The Lancet. Digital health·2024

Same author

Multi-domain improves classification in out-of-distribution and data-limited scenarios for medical image analysis.

Scientific reports·2024

Same author

Transformer Module Networks for Systematic Generalization in Visual Question Answering.

IEEE transactions on pattern analysis and machine intelligence·2024

Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026

Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026

Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026

Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026

Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026

Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026

See all related articles

This summary is machine-generated.

Deep neural networks (DNNs) can solve the insideness problem in image segmentation. However, recurrent networks trained on small images generalize better for this task, handling long-range spatial dependencies effectively.

Area of Science:

Computer Vision
Artificial Intelligence
Machine Learning

Background:

The insideness problem in image segmentation determines if pixels are inside or outside a region.
Deep neural networks (DNNs) are powerful for segmentation but their capability for long-range spatial dependencies in insideness is unclear.
Analyzing insideness in isolation, without texture or semantic cues, isolates the core challenge.

Purpose of the Study:

To analyze the insideness problem in isolation.
To evaluate the ability of DNNs to solve the insideness problem.
To identify network architectures that generalize well for insideness detection.

Main Methods:

Isolated the insideness problem from other segmentation cues.
Tested DNNs with varying complexity for solving the insideness problem.

More Related Videos

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Oct 23, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

Investigated recurrent networks trained on small images for generalization capabilities.

Main Results:

DNNs with few units possess the complexity to solve the insideness problem for any curve.
Standard DNNs struggle to learn general solutions for the insideness problem.
Recurrent networks trained on small images demonstrate strong generalization for insideness detection.

Conclusions:

Recurrent networks can effectively handle long-range dependencies by decomposing them into local operations.
Training recurrent networks on smaller images mitigates difficulties associated with long unrolling steps.
Recurrent networks offer a promising approach for robust insideness problem solutions in image segmentation.