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 Video

Updated: Oct 30, 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

Emotion Recognition on Edge Devices: Training and Deployment.

Vlad Pandelea¹, Edoardo Ragusa², Tommaso Apicella²

¹School of Computer Science and Engineering, Nanyang Technological University, 50 Nanyang Ave, Singapore 639798, Singapore.

Sensors (Basel, Switzerland)

|July 2, 2021

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

A Two-Dimensional (2-D) Sensor Network Architecture with Artificial Intelligence Models for the Detection of Magnetic Anomalies.

Sensors (Basel, Switzerland)·2026

Same author

Survey on sentiment analysis: evolution of research methods and topics.

Artificial intelligence review·2023

Same author

From a Chemotherapeutic Drug to a High-Performance Nanocatalyst: A Fast Colorimetric Test for Cisplatin Detection at ppb Level.

Biosensors·2022

Same author

Enhancing Cyber Security of LoRaWAN Gateways under Adversarial Attacks.

Sensors (Basel, Switzerland)·2022

Same author

Mood of the Planet: Challenging Visions of Big Data in the Arts.

Cognitive computation·2021

Same author

T-RexNet-A Hardware-Aware Neural Network for Real-Time Detection of Small Moving Objects.

Sensors (Basel, Switzerland)·2021

Large transformer models combined with linear classifiers offer efficient emotion recognition on edge devices. This approach achieves competitive performance with real-time inference and fast training, overcoming computational cost challenges.

Area of Science:

Natural Language Processing
Machine Learning
Computer Vision

Background:

Large transformer models excel in natural language processing (NLP) tasks like emotion recognition.
Deploying these computationally intensive models on resource-constrained edge devices presents a significant challenge.

Purpose of the Study:

To investigate efficient deployment of large transformer models for emotion recognition on edge devices.
To achieve competitive performance with real-time inference and fast training.

Main Methods:

Utilizing large transformers as feature extractors combined with hardware-friendly linear classifiers.
Analyzing batch and Online Sequential Learning methods.
Applying dimensionality reduction and pre-training techniques.

Keywords:

deep learning embedded systems emotion recognition

More Related Videos

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

Exploring the Use of Isolated Expressions and Film Clips to Evaluate Emotion Recognition by People with Traumatic Brain Injury

Exploring the Use of Isolated Expressions and Film Clips to Evaluate Emotion Recognition by People with Traumatic Brain Injury

Published on: May 15, 2016

Related Experiment Videos

Last Updated: Oct 30, 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

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

Exploring the Use of Isolated Expressions and Film Clips to Evaluate Emotion Recognition by People with Traumatic Brain Injury

Exploring the Use of Isolated Expressions and Film Clips to Evaluate Emotion Recognition by People with Traumatic Brain Injury

Published on: May 15, 2016

Main Results:

The combined approach achieves competitive performance for emotion recognition.
Real-time inference and fast training are enabled on edge devices.
Dimensionality reduction and pre-training further improve latency and performance.

Conclusions:

Large transformers and linear classifiers offer an effective solution for efficient emotion recognition on edge devices.
The proposed system demonstrates practical implementation on edge accelerators and smartphones.
This method balances high performance with the computational constraints of edge computing.