Journal of Applied Science and Engineering

Published by Tamkang University Press

ESCI jase impact factor scopus logo open access rate of Scopus journal

Interpretable Artificial Intelligence in Emotional Modeling of Digital Media Characters

 2026

 Volume 35

Jingyi Yao

School of Art, Sanjiang University, Nanjing, Jiangsu, 210012, China

Received: June 05, 2026
Accepted: August 01, 2026
Publication Date: September 18, 2026

上傳圖片

Comparative Analysis of the Emotional Detection Models (a) Accuracy, (b) Precision, (c) Recall, and (d) F1-Score

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: BibTeX | http://dx.doi.org/10.6180/jase.202612_35.055

Download PDF

The animation industry has emerged as a dominant force in digital entertainment, encompassing gaming, virtual storytelling, and interactive media. A crucial component in this domain is accurate and interpretable emotional modeling of stylized characters. Although various deep learning (DL) models have been effectively applied to human emotion recognition, the task of emotional modeling in non-human, digitally stylized characters remain underexplored, particularly from an explainable AI (XAI) perspective. This research presents the Intelligent Coral Reefs Optimizer-tuned Attention-enriched MobileNetV2 (ICR-Att-MobileNetV2) framework for interpretable emotion recognition in digital media characters. A custom Detect Emotions of Your Favorite Toons (DEYFT) dataset was collected with annotated Tom and Jerry cartoon images depicting five core emotions: Angry, Happy, Sad, Surprised, and Unknown. Images were preprocessed via resizing and pixel normalization. The proposed model incorporates U-Net for precise facial segmentation and applies an Intelligent Coral Reefs Optimizer (ICRO)-tuned MobileNetV2 with attention mechanisms for emotion classification. Implemented in Python using TensorFlow, the model achieved a higher classification accuracy (99.84%), Precision, Recall, and F1-Score, which demonstrates the strong generalization and robustness on stylized facial data. Furthermore, attention heatmaps provided visual interpretability, aiding animators and developers in understanding emotion localization. This research highlights the potential of interpretable DL in advancing animation design, emotionally intelligent virtual agents, and automated content generation within creative media ecosystems.

Keywords: Emotional Modeling, Artificial Intelligence (AI), Digital Media Characters, Emotion Classification, Intelligent Coral Reefs Optimizer-tuned Attention-enriched MobileNetV2 (ICR-Att-MobileNetV2)

  1. [1] M. Zhao, (2023) “The Emotion Recognition in Psychology of Human-robot Interaction” Psychomachina 1: 1-11. DOI: https://doi.org/10.59388/pm00331.
  2. [2] L. Meng, C. Tang, and X. Yang, (2025) “Evaluating the Emotional Impact of Virtual Expressions: ERPs Analysis of Smart Agent Appearance” Behaviour & Information Technology 45(5): DOI: https://doi.org/10.1080/0144929X.2025.2539277.
  3. [3] S. Zhang, X. Liu, X. Yang, Y. Shu, N. Liu, D. Zhang, and Y. J. Liu, (2021) “The Influence of Key Facial Features on Recognition of Emotion in Cartoon Faces” Frontiers in Psychology 12: 687974. DOI: https://doi.org/10.3389/fpsyg.2021.687974.
  4. [4] X. Wang and X. Qiu, (2024) “The Positive Effect of Artificial Intelligence Technology Transparency on Digital Endorsers: Based on the Theory of Mind Perception” Journal of Retailing and Consumer Services 78: 103777. DOI: https://doi.org/10.1016/j.jretconser.2024.103777.
  5. [5] A. Ashari, L. W. Jo, J. Peh, and G. Oliver, (2025) “Facial Features of Cartoon Characters and Their Perceived Attributes” Scientific Reports 15(1): 1093. DOI: https://doi.org/10.1038/s41598-024-82677-0.
  6. [6] S. Sonlu, K. Shabanovi, U. Güdükbay, and F. Durupinar. “Emotional Expression as a Means of Communicating Virtual Human Personalities”. In: Modeling Visual Aesthetics, Emotion, and Artistic Style. Cham: Springer International Publishing, 2023, 293-312. DOI: https://doi.org/10.1007/978-3-031-50269-9_17.
  7. [7] X. Su and H. G. Kim, (2024) “Automatic Stylized Action Generation in Animation Using Deep Learning” IEEE Access: DOI: https://doi.org/10.1109/ACCESS.2024.3486024.
  8. [8] N. Jain, V. Gupta, S. Shubham, A. Madan, A. Chaudhary, and K. C. Santosh, (2022) “Understanding Cartoon Emotion Using Integrated Deep Neural Network on Large Dataset” Neural Computing and Applications 34(24): 21481-21501. DOI: https://doi.org/10.1007/s00521-021-06003-9.
  9. [9] A. Madan, J. Parikh, R. Jain, A. Gupta, A. Chaudhary, D. Chadha, and S. Shubham, (2023) “Enhanced Deep Learning Network for Emotion Recognition from GIF” Intelligent Decision Technologies 17(2): 415-433. DOI: https://doi.org/10.3233/IDT-220158.
  10. [10] S. AlZu’bi, R. Abu Zitar, B. Hawashin, S. Abu Shanab, A. Zraiqat, A. Mughaid, K. H. Almotairi, and L. Abualigah, (2022) “A Novel Deep Learning Technique for Detecting Emotional Impact in Online Education” Electronics 11(18): 2964. DOI: https://doi.org/10.3390/electronics11182964.
  11. [11] M. Y. Chuttur and A. Nazurally, (2022) “A Multi-Modal Approach to Detect Inappropriate Cartoon Video Contents Using Deep Learning Networks” Multimedia Tools and Applications 81(12): 16881-16900. DOI: https://doi.org/10.1007/s11042-022-12709-2.
  12. [12] R. Sharma and V. Kukreja, (2025) “Deciphering Emotions in Comics: Analysis of Emotion Classification of Comic Characters via Attention-Based Deep Learning Models” Multimedia Tools and Applications: 1-33. DOI: https://doi.org/10.1007/s11042-025-21017-4.
  13. [13] B. Zhang, (2025) “Development and Implementation of a Deep Learning-Based Smart Evaluation System for Animated Films” Journal of Computational Methods in Sciences and Engineering 25(3): 2529-2539. DOI: https://doi.org/10.1177/14727978251314558.
  14. [14] A. Dutta, S. Biswas, and A. K. Das, (2024) “EmoComicNet: A Multi-Task Model for Comic Emotion Recognition” Pattern Recognition 150: 110261. DOI: https://doi.org/10.1016/j.patcog.2024.110261.
  15. [15] S. Liang, (2023) “Analysis of Emotional Deconstruction and the Role of Emotional Value for Learners in Animation Works Based on Digital Multimedia Technology” IET Software 2023(1): 5566781. DOI: https://doi.org/10.1049/2023/5566781.