Search research articles

Search research articles

Related Experiment Videos

Rotation recovery from spherical images without correspondences.

Ameesh Makadia¹, Kostas Daniilidis

¹GRASP Laboratory, Department of Computer and Information Science, University of Pennsylvania, 3330 Walnut Street, Philadelphia, PA 19104, USA. makadia@cis.upenn.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence

|June 24, 2006

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

Light Field Neural Rendering.

IEEE transactions on pattern analysis and machine intelligence·2023

Same author

3D Bird Reconstruction: a Dataset, Model, and Shape Recovery from a Single View.

Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision·2022

Same author

Event-Based Vision: A Survey.

IEEE transactions on pattern analysis and machine intelligence·2020

Same author

MonoCap: Monocular Human Motion Capture using a CNN Coupled with a Geometric Prior.

IEEE transactions on pattern analysis and machine intelligence·2018

Same author

Sparse Representation for 3D Shape Estimation: A Convex Relaxation Approach.

IEEE transactions on pattern analysis and machine intelligence·2017

Same author

Scale Space for Camera Invariant Features.

IEEE transactions on pattern analysis and machine intelligence·2015

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

This study presents a novel method for estimating large rotations directly from spherical images without needing correspondences. The technique leverages spherical harmonic coefficients for accurate 3D shape alignment.

Area of Science:

Computer Vision
Geometry Processing
Signal Processing

Background:

Estimating rotations from spherical images is crucial for 3D shape alignment.
Existing methods often struggle with large rotations or require point correspondences.

Purpose of the Study:

To develop a method for direct rotation estimation from spherical images without correspondences.
To enable accurate alignment of large rotations and impact 3D shape alignment.

Main Methods:

Utilizing the unitary mapping property of spherical harmonic coefficients under rotation.
Employing the SO(3)-Fourier transform of image correlation for rotation estimation.
Implementing a direct search in a discretized rotation space based on harmonic expansion bandwidth.

Related Experiment Videos

Refining rotation estimates using a novel decoupling of the rotational shift theorem with respect to Euler angles.

Main Results:

Demonstrated suitability of the method for estimating large rotations.
Showcased the dependence of rotation estimation accuracy on bandwidth and spherical harmonic coefficient selection.
Validated the effectiveness of the iterative refinement scheme.

Conclusions:

The proposed method offers a robust solution for rotation estimation directly from spherical images.
The technique shows significant potential for applications in 3D shape alignment and related fields.
Further research can explore optimal bandwidth selection and coefficient choices for enhanced performance.