What question did this study set out to answer?

This research aims to improve facial emotion recognition (FER) accuracy by utilizing a deep convolutional neural network (DCNN) with transfer learning.

April 27, 2021ElectronicsOpen Access

Facial Emotion Recognition Using Transfer Learning in the Deep CNN

View Full Paper

Ask AI

Bookmark

Authors

MAM. A. H. AkhandUniversity of Fukui SRShuvendu RoyQueen's University NSNazmul SiddiqueUniversity of Ulster

Discussion

Member takes

Implication

Randomized trial demonstrates enhanced facial emotion recognition accuracy using deep CNN and transfer learning.

Key Points

This research aims to improve facial emotion recognition (FER) accuracy by utilizing a deep convolutional neural network (DCNN) with transfer learning.
Developed a deep CNN model for FER using transfer learning techniques.
Fine-tuned the model by adjusting pre-trained DCNN layers based on facial emotion data.
Tested the method on two datasets, KDEF and JAFFE, incorporating both frontal and profile images.
Achieved FER accuracy of 96.51% on the KDEF dataset and 99.52% on the JAFFE dataset using DenseNet-161.
Showed significant improvement over existing FER methods, especially with profile views on the KDEF dataset.
Validated the proposed approach with 10-fold cross-validation, reflecting robust performance.

Cite This Study

Akhand et al. (2021) studied this question.

synapsesocial.com/papers/6a00ca85ef8139f8ff77a850 https://doi.org/10.3390/electronics10091036

View Full Paper

Ask AI

Bookmark

Facial Emotion Recognition Using Transfer Learning in the Deep CNN

Discussion

Member takes

Implication

Key Points

Cite This Study

Facial Emotion Recognition Using Transfer Learning in the Deep CNN

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider