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

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

Social Attention in Electronic Picture Books with Social Scenes for Children with Autism Spectrum Disorder: Insights from Eye-Tracking Studies.

Behavioral sciences (Basel, Switzerland)·2026

Same author

A probabilistic assessment of dietary heavy metal exposure and its temporal trends in Chongqing China from 2012 to 2022.

Scientific reports·2026

Same author

Genotoxicity mode of action exploration of polyethylene glycol terephthalate (PET) acetic acid migration solution under repeated-dose exposure in rats.

Mutation research. Genetic toxicology and environmental mutagenesis·2025

Same author

[Assessment of Dietary Arsenic Exposure Levels and the Associated Health Risks in Chongqing City, China].

Sichuan da xue xue bao. Yi xue ban = Journal of Sichuan University. Medical science edition·2025

Same author

Probabilistic assessment of the cumulative risk from dietary heavy metal exposure in Chongqing, China using a hazard-driven approach.

Scientific reports·2025

Same author

Genotoxic mode of action and threshold exploration of 2-methyl furan under 120-day sub-chronic exposure in male Sprague-Dawley rats.

Ecotoxicology and environmental safety·2024

Search research articles

Related Experiment Video

Updated: Nov 10, 2025

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

Head Pose Estimation through Keypoints Matching between Reconstructed 3D Face Model and 2D Image.

Leyuan Liu^1,2, Zeran Ke¹, Jiao Huo¹

¹National Engineering Research Center for E-Learning, Central China Normal University, Wuhan 430079, China.

Sensors (Basel, Switzerland)

|April 3, 2021

Summary

This study introduces a novel head pose estimation method that bypasses the need for labeled training data. By matching keypoints between a reconstructed 3D face model and 2D images, it achieves high accuracy across multiple datasets.

Keywords:

3D face reconstruction computer vision facial keypoints matching head pose estimation

More Related Videos

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

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

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

Published on: October 18, 2024

Related Experiment Videos

Last Updated: Nov 10, 2025

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

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

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

Published on: October 18, 2024

Area of Science:

Computer Vision
Machine Learning
3D Reconstruction

Background:

Head pose estimation typically relies on supervised learning with accurate ground-truth labels.
Acquiring precise head pose labels is challenging due to equipment and methodological limitations.

Purpose of the Study:

To propose a novel head pose estimation method that does not require labeled training data.
To enable accurate head pose estimation through 3D-2D keypoint matching.

Main Methods:

Reconstructs a personalized 3D face model from input head images using convolutional neural networks.
Employs an iterative optimization algorithm for efficient 3D-2D keypoint matching under perspective transformation constraints.

Main Results:

Achieves excellent cross-dataset performance on five benchmark datasets (Pointing'04, BIWI, AFLW2000, Multi-PIE, Pandora).
Surpasses most existing state-of-the-art approaches, demonstrating robust head pose estimation without dataset-specific training.

Conclusions:

The proposed method offers a viable, label-free alternative for head pose estimation.
Its effectiveness is validated by superior performance across diverse datasets, highlighting its generalizability.