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 Videos

Neural computation approach for developing a 3D shape reconstruction model.

S Y Cho¹, T S Chow

¹Department of Electronic Engineering, City University of Hong Kong, Kowloon Tong, Hong Kong. eedsycho@cityu.edu.hk

IEEE Transactions on Neural Networks

|February 6, 2008

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

Validating pelvic stone angle as a substitute for infundibulopelvic angle in RIRS outcome prediction.

Journal of the Formosan Medical Association = Taiwan yi zhi·2025

Same author

Polydactyly and syndactyly linked to GLI3 and TBX5 mutations: A pediatric case report.

Global medical genetics·2025

Same author

Effect of a latent tuberculosis infection programme for healthcare workers in a country with an intermediate tuberculosis burden.

The Journal of hospital infection·2023

Same author

Development of machine learning models for the surveillance of colon surgical site infections.

The Journal of hospital infection·2023

Same author

Efflux pump gene expression study using RNA-seq in multidrug-resistant TB.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2021

Same author

Filariasis in axillary lymph node.

Tropical biomedicine·2021

Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013

Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011

Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011

Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011

Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011

Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011

See all related articles

This study introduces a novel neural network for 3D shape reconstruction, overcoming limitations of traditional shape from shading models by learning complex reflectance properties from images.

Area of Science:

Computer Vision
Machine Learning
3D Reconstruction

Background:

Traditional shape from shading models struggle with specular reflections and unknown surface reflectivity.
Real-world images often deviate from simplified Lambertian or purely specular models.

Purpose of the Study:

To propose a new neural-based 3D shape reconstruction model that addresses limitations of existing methods.
To develop and evaluate novel neural reflectance models for improved accuracy.

Main Methods:

Introduced feedforward neural network (FNN) for diffuse reflectance generalization.
Presented radial basis function (RBF) model for specular reflectance generalization.
Developed a hybrid FNN-RBF model for complex surface reflectance.

Related Experiment Videos

Main Results:

Demonstrated effective 3D shape reconstruction using the proposed neural models.
Validated performance across synthetic and real images with varying specularities.
Showcased robustness under unknown illumination and noise conditions.

Conclusions:

The proposed neural reflectance models significantly improve 3D shape reconstruction accuracy.
The hybrid model effectively captures both diffuse and specular surface properties.
This approach offers a robust solution for shape from shading in complex real-world scenarios.