ItemArticle

Actor-critic guided CDBN with GAN augmentation for robust facial emotion recognition

Title
Actor-critic guided CDBN with GAN augmentation for robust facial emotion recognition
Creator
S, Akshay; S R, Jnana Sai; B R, Sinchana; M, Kannan; Mukhopadhyay, Adwitiya
Description
Facial emotion recognition (FER) remains challenging under limited data, noise, and occlusion. This study introduces an ActorCritic Convolutional Deep Belief Network (ACCDBN) that unifies Generative Adversarial Network (GAN)based augmentation, deep probabilistic feature learning, and reinforcement-driven optimization. Conditional GANs expand minority emotion classes, enhancing data diversity, while the CDBN extracts hierarchical texture features through convolutional and restricted Boltzmann layers. An ActorCritic module dynamically refines representations by rewarding accurate emotion classification and penalizing uncertain predictions. Trained and validated on the CK+ dataset with five-fold cross-validation, the proposed model achieves higher accuracy and stability than CNN, LSTM, and ResNet-50 baselines, maintaining strong performance under noise and occlusion. The approach demonstrates how reinforcement-guided generative learning can improve both accuracy and robustness in FER tasks.1. To implement this, the research utilised the publicly available Cohn-Kanade+ dataset, consisting of eight classes with samples of 920 grey-scale images.2. An improved ACCDBN model outperformed with 90.4% accuracy and 0.69 MCC (Mathews Correlation Coefficient) in 5-fold cross-validation using the cGAN-generated dataset and 87% on the CK+ dataset3. The main objective is to present an advanced facial emotion recognition (FER) system that combines a Convolution Deep Belief Network (CDBN) with a model-free reinforcement learning technique, namely the actor-critic approach. 2025 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license. http://creativecommons.org/licenses/by-nc-nd/4.0/
Source
MethodsX;Volume;16;Issue;;Article No.;103774;
Date
01-01-2026
Publisher
Elsevier B.V.
Subject
Conditional generative adversarial network; Convolutional deep belief network; Face emotion recognition; Improved ACCDBN; Model-free reinforcement learning (actor-critic)
Coverage
S A., Department of Computer Science, School of Computing, Amrita Vishwa Vidyapeetham, Mysuru Campus, Karnataka, India; S R J.S., Department of Computer Science, School of Computing, Amrita Vishwa Vidyapeetham, Mysuru Campus, Karnataka, India; B R S., Department of Computer Science, School of Computing, Amrita Vishwa Vidyapeetham, Mysuru Campus, Karnataka, India; M K., Department of Computer Science, School of Sciences, Christ University, Karnataka, Bengaluru, India; Mukhopadhyay A., Department of Computer Science, School of Computing, Amrita Vishwa Vidyapeetham, Mysuru Campus, Karnataka, India
Rights
All Open Access; Gold Open Access; Green Open Access
Relation
ISSN: 22150161;
Format
online
Language
English
Type
Article

Citation

S, Akshay; S R, Jnana Sai; B R, Sinchana; M, Kannan; Mukhopadhyay, Adwitiya, “Actor-critic guided CDBN with GAN augmentation for robust facial emotion recognition,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 18, 2026, https://archives.christuniversity.in/items/show/22390.

Output Formats