Author:
Dr. Sheshang Degadwala ,Radhika Thakkar
Abstract
Deception detection through facial and audio transcript features has gained traction due to its potential in enhancing security and communication integrity. This review aims to consolidate existing research on leveraging facial and audio features for identifying deceptive behavior. The motivation behind this study is the increasing demand for reliable deception detection mechanisms in various domains, including security and psychology. Despite advancements, limitations persist in achieving high accuracy across diverse contexts and individual differences. The objective of this review is to evaluate the effectiveness of different methods used in detecting deception from facial expressions and audio cues, identifying strengths and weaknesses of each approach, and suggesting future directions for improving accuracy through advanced techniques.
Reference64 articles.
1. Dinges, Laslo, et al. “Exploring Facial Cues: Automated Deception Detection Using Artificial Intelligence.” Neural Computing and Applications, vol. 0123456789, 2024, https://doi.org/10.1007/s00521-024-09811-x.
2. Prome, Shanjita Akter, et al. “Deception Detection Using Machine Learning (ML) and Deep Learning (DL) Techniques: A Systematic Review.” Natural Language Processing Journal, vol. 6, no. October 2023, 2024, p. 100057, https://doi.org/10.1016/j.nlp.2024.100057.
3. Abdulridha, Fahad, and Baraa M. Albaker. “Non-Invasive Real-Time Multimodal Deception Detection Using Machine Learning and Parallel Computing Techniques.” Social Network Analysis and Mining, vol. 14, no. 1, 2024, https://doi.org/10.1007/s13278-024-01255-4.
4. Bahaa, Mohamed, et al. “Advancing Automated Deception Detection: A Multimodal Approach to Feature Extraction and Analysis.” Computer Vision and Pattern Recognition, July 2024, pp. 1–13, https://doi.org/10.48550/arXiv.2407.06005.
5. Guo, Xiaobao, et al. “Benchmarking Cross-Domain Audio-Visual Deception Detection.” IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, 2024, pp. 1–10, http://arxiv.org/abs/2405.06995.