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

Vector Transformation in Rotating Coordinate Systems01:16

Vector Transformation in Rotating Coordinate Systems

2.3K
Consider a vector rotating about an axis with an angular velocity, such that its tip sweeps a circular path.
2.3K
Curvilinear Motion: Rectangular Components01:23

Curvilinear Motion: Rectangular Components

973
Curvilinear motion characterizes the movement of a particle or object along a curved path, notably evident when envisioning a car navigating a winding road. If the car starts at point A, its position vector is established within a fixed frame of reference, where the ratio of the position vector to its magnitude signifies the unit vector pointing in the position vector's direction.
As the car advances, its position evolves over time. Quantifying the car's velocity involves computing the...
973
Gauss's Law: Cylindrical Symmetry01:20

Gauss's Law: Cylindrical Symmetry

9.0K
A charge distribution has cylindrical symmetry if the charge density depends only upon the distance from the axis of the cylinder and does not vary along the axis or with the direction about the axis. In other words, if a system varies if it is rotated around the axis or shifted along the axis, it does not have cylindrical symmetry. In real systems, we do not have infinite cylinders; however, if the cylindrical object is considerably longer than the radius from it that we are interested in,...
9.0K
Convolution Properties I01:20

Convolution Properties I

445
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
445
Curvilinear Motion: Normal and Tangential Components01:27

Curvilinear Motion: Normal and Tangential Components

708
When a car traverses a curved road, its motion can be elucidated by breaking it down into tangential and normal components. The car-centric coordinates attached to the vehicle move with it.
The positive direction of the t-axis aligns with the increasing position of the car along the curved path, denoted by the unit vector ut. Simultaneously, the n-axis, perpendicular to the t-axis, dissects the curved path into differential arc segments, each forming the arc of a circle with a radius of...
708
Rotation with Constant Angular Acceleration - II01:16

Rotation with Constant Angular Acceleration - II

6.8K
Kinematics is the description of motion. The kinematics of rotational motion discusses the relationships between rotation angle, angular velocity, angular acceleration, and time. One can describe many things with great precision using kinematics, but kinematics does not consider causes. For example, a large angular acceleration describes a very rapid change in angular velocity without any consideration of its cause. Thus, rotational kinematics does not represent the laws of nature.
The first...
6.8K

You might also read

Related Articles

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

Sort by
Same author

Postsurgical detection of glioma recurrence using MRI radiomics.

Neuro-oncology advances·2026
Same author

A comparative study of deep learning for cortical lesion MRI segmentation with explainability analysis in multiple sclerosis.

NeuroImage. Clinical·2026
Same author

Fully automatic left ventricle segmentation in [Formula: see text]Rb PET/CT Using a semi-supervised nnU-net.

EJNMMI research·2026
Same author

Impact of CT dose on AI performance: A comparison of radiomics, deep, and foundation models in a multicentric anthropomorphic phantom study.

Medical physics·2026
Same author

A multi-modal deep learning network for the classification of paramagnetic rim and remyelinated lesions in multiple sclerosis.

Multiple sclerosis (Houndmills, Basingstoke, England)·2026
Same author

Instance-level quantitative saliency in multiple sclerosis lesion segmentation.

Scientific reports·2026

Related Experiment Video

Updated: Dec 16, 2025

Author Spotlight: Insights into the Analysis of Human Interaction with 3D Virtual Objects
06:36

Author Spotlight: Insights into the Analysis of Human Interaction with 3D Virtual Objects

Published on: October 18, 2024

1.3K

Local rotation invariance in 3D CNNs.

Vincent Andrearczyk1, Julien Fageot2, Valentin Oreiller3

  • 1Institute of Information Systems, University of Applied Sciences Western Switzerland (HES-SO), Sierre, Switzerland.

Medical Image Analysis
|July 6, 2020
PubMed
Summary

Locally Rotation Invariant (LRI) image analysis enables deep learning models to learn from images with arbitrary orientations. This approach significantly reduces trainable parameters and data requirements, outperforming standard 3D CNNs.

Keywords:
3D TextureConvolutional neural networkLocal rotation invarianceSteerable filters

More Related Videos

Three Dimensional Vestibular Ocular Reflex Testing Using a Six Degrees of Freedom Motion Platform
10:12

Three Dimensional Vestibular Ocular Reflex Testing Using a Six Degrees of Freedom Motion Platform

Published on: May 23, 2013

16.2K
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
07:05

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine

Published on: October 27, 2016

9.5K

Related Experiment Videos

Last Updated: Dec 16, 2025

Author Spotlight: Insights into the Analysis of Human Interaction with 3D Virtual Objects
06:36

Author Spotlight: Insights into the Analysis of Human Interaction with 3D Virtual Objects

Published on: October 18, 2024

1.3K
Three Dimensional Vestibular Ocular Reflex Testing Using a Six Degrees of Freedom Motion Platform
10:12

Three Dimensional Vestibular Ocular Reflex Testing Using a Six Degrees of Freedom Motion Platform

Published on: May 23, 2013

16.2K
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
07:05

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine

Published on: October 27, 2016

9.5K

Area of Science:

  • Computer Vision
  • Medical Imaging
  • Deep Learning

Background:

  • Locally Rotation Invariant (LRI) image analysis is crucial for applications with arbitrary local structure orientations, especially in medical imaging.
  • While globally rotation invariant Convolutional Neural Networks (CNNs) exist, LRI in deep learning remains underexplored.
  • Existing LRI methods like Local Binary Patterns (LBP) and steerable filterbanks are foundational in texture analysis.

Purpose of the Study:

  • To propose and compare novel methods for achieving Locally Rotation Invariant (LRI) Convolutional Neural Networks (CNNs) with directional sensitivity.
  • To investigate the effectiveness of LRI CNNs in reducing trainable parameters and data needs compared to standard 3D CNNs.
  • To evaluate LRI CNNs on 3D datasets, including synthetic textures and pulmonary nodule classification in CT scans.

Main Methods:

  • Developed LRI CNNs using orientation channels derived from explicitly rotated kernels or steerable filters.
  • Employed steerable filters with solid Spherical Harmonics (SH) for efficient 3D rotation sampling and parameter reduction.
  • Investigated a third strategy for LRI using rotational invariants from learned solid SH responses.

Main Results:

  • LRI CNNs demonstrated superior performance compared to standard 3D CNNs on 3D datasets, including synthetic rotated patterns and pulmonary nodule classification.
  • The proposed LRI methods achieved a significant reduction in trainable parameters and training data requirements.
  • The study confirmed the importance of LRI image analysis for handling rotational variations in deep learning.

Conclusions:

  • Locally Rotation Invariant (LRI) CNNs offer a powerful approach for analyzing images with arbitrary orientations, particularly in medical imaging.
  • LRI CNNs drastically reduce model complexity and data needs while achieving state-of-the-art performance.
  • The proposed methods, leveraging orientation channels and Spherical Harmonics, provide efficient and effective solutions for rotation-invariant deep learning.