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

Phytochemical Characterization and Neuropharmacological Assessment of <i>Combretum indicum</i> Methanolic Extract With Integrated Molecular Docking and Dynamics Simulation for Anxiety and Depression Therapy.

Food science & nutrition·2026

Same author

The genetic potential of onion (Allium cepa L.) germplasm for enhancing productivity, adaptability, and sustainability in the current era of climate change.

Journal of the science of food and agriculture·2026

Same author

Impact of long-term straw and manure incorporation on carbon sequestration and yield through alteration of aluminum and iron oxides in acidic red soil.

Scientific reports·2026

Same author

Screen use and sleep in early childhood: a national survey of Australian families.

BMC public health·2026

Same author

Efficient Recovery of Collagen from Tannery Waste Materials and Its Integration into Functional Hydrogel Systems.

Gels (Basel, Switzerland)·2026

Same author

Mechanistic insights into methyl violet dye degradation using fish bone-derived hydroxyapatite: LC-MS/MS identification and computational analysis.

Journal of environmental management·2026

Search research articles

Related Experiment Video

Updated: Dec 31, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Trajectory-Based Air-Writing Recognition Using Deep Neural Network and Depth Sensor.

Md Shahinur Alam¹, Ki-Chul Kwon¹, Md Ashraful Alam²

¹Department of Computer and Communication Engineering, Chungbuk National University, Cheongju, Chungbuk 28644, Korea.

Sensors (Basel, Switzerland)

|January 16, 2020

Summary

This study introduces an advanced air-writing recognition system using 3D trajectories and deep learning (LSTM and CNN). The system achieves state-of-the-art 99.32% accuracy for recognizing characters and digits, addressing data limitations.

Keywords:

CNN LSTM TOF camera air-writing depth sensor digit recognition human-computer interaction trajectory recognition

More Related Videos

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

Related Experiment Videos

Last Updated: Dec 31, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

Area of Science:

Human-Computer Interaction
Artificial Intelligence
Computer Vision

Background:

Trajectory-based writing systems offer advantages over traditional methods in specific applications.
Recognizing free-space handwriting is challenging due to variations in character shapes and writing styles.

Purpose of the Study:

To develop a robust air-writing recognition system using 3D trajectory data.
To improve feature selection and normalization techniques for trajectory data.
To establish a new benchmark accuracy for air-writing recognition.

Main Methods:

Utilized a depth camera to capture fingertip trajectories in 3D.
Applied nearest neighbor and root point translation for trajectory normalization.
Employed Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) for character recognition.

Main Results:

Achieved 99.32% accuracy on the 6D motion gesture (6DMG) alphanumeric character dataset, surpassing previous benchmarks.
Demonstrated model invariance for both digits and characters.
Published a new dataset of 21,000 digits to support future research.

Conclusions:

The proposed air-writing recognition system demonstrates high accuracy and robustness.
The developed model effectively handles variations in writing styles and character forms.
The released dataset will aid in advancing trajectory-based writing system research.