Mudunuru Suneel; Banothu Yedukondala Venkata Naga Raja Swamy; Madhava Rao Maganti; Seva Sreedhar Babu; P. Rama Koteswara Rao; Vijaya Kumari Devarapalli; Kama Ramudu

A Lightweight Face Anti-Spoofing Framework with Spoof Artifact Enhancement and Adaptive Feature Fusion

PDF (1460KB), PP.384-410

Views: 0 Downloads: 0

Author(s)

Mudunuru Suneel ^1,* Banothu Yedukondala Venkata Naga Raja Swamy ² Madhava Rao Maganti ³ Seva Sreedhar Babu ⁴ P. Rama Koteswara Rao ⁵ Vijaya Kumari Devarapalli ³ Kama Ramudu ¹

1. Department of Electronics and Communication Engineering, Aditya University, Surampalem, Andhra Pradesh, India

2. Department of Electronics and Communication Engineering, Lakireddy Bali Reddy College of Engineering, Mylavaram, Andhra Pradesh, India

3. Department of Computer Science and Engineering, Vijaya Institute of Technology for Women, Enikepadu, Vijayawada, Andhra Pradesh, India

4. Department of Electronics and Communication Engineering, Sree Vahini Institute of Science and Technology, Thiruvuru, Andhra Pradesh, India

5. Department of Computer Science and Engineering, Sree Dattha Institute of Engineering & Science, Sheriguda, Hyderabad, Telangana, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijem.2026.05.21

Received: 14 Jul. 2026 / Revised: 17 Aug. 2026 / Accepted: 7 Sep. 2026 / Published: 8 Oct. 2026

Index Terms

Biometric Security, Face Anti-Spoofing, Presentation Attack Detection, Synthetic Multi-Modal Representation, Spoof Artifact Enhancement, EfficientNetV2, Cross-Modal Consistency Learning, Adaptive Feature Fusion

Abstract

Face recognition is widely used for biometric authentication in applications such as mobile devices, financial services, intelligent surveillance, access control, and border security. However, face recognition systems remain vulnerable to presentation attacks, including printed photographs, replay attacks, and three-dimensional masks. Although recent deep learning-based face anti-spoofing (FAS) methods have achieved substantial improvements, many existing approaches still involve considerable computational cost, provide limited emphasis on fine-grained spoof artifacts, and face challenges in effectively integrating heterogeneous representations. To address these limitations, this paper proposes
a Lightweight Face Anti-Spoofing Framework with Synthetic Multi-Modal Representation, Spoof Artifact Enhancement, and Adaptive Feature Fusion. The framework starts from a single RGB facial image and constructs complementary Depth and Near-Infrared (NIR) representations using a Depth and Near-Infrared Construction Module (DNCM), rather than requiring dedicated Depth or NIR sensors. A shared EfficientNetV2 backbone is then employed to extract features from the three representations with reduced computational redundancy. The proposed Spoof Artifact Enhancement Module (SAEM) emphasizes subtle spoof-specific visual cues, while Cross-Modal Consistency Learning (CMCL) reduces representation discrepancies across the constructed modalities. Subsequently, the Adaptive Feature Fusion Module (AFFM) dynamically weights the refined representations according to their discriminative contribution. Extensive experiments on CelebA-Spoof, CASIA-SURF, HQ-WMCA, and SiW-M demonstrate the effectiveness of the proposed framework. The framework achieves accuracies of 98.16%, 97.54%, 99.08%, and 97.18%, with corresponding ACER values of 2.87%, 4.36%, 2.03%, and 4.18%, respectively. Across five independent runs, statistical analysis further indicates consistent performance with significant improvements over the selected baseline. In addition, the framework requires only 9.8 million parameters and 2.1 GFLOPs and achieves an inference speed of 82 FPS, demonstrating a favourable balance between detection effectiveness and computational efficiency. The results indicate that synthetic multi-modal representation combined with explicit spoof artifact enhancement and adaptive feature fusion can provide an efficient solution for robust and real-time face anti-spoofing.

Cite This Paper

Mudunuru Suneel, Banothu Yedukondala Venkata Naga Raja Swamy, Madhava Rao Maganti, Seva Sreedhar Babu, P. Rama Koteswara Rao, Vijaya Kumari Devarapalli, Kama Ramudu, "A Lightweight Face Anti-Spoofing Framework with Spoof Artifact Enhancement and Adaptive Feature Fusion", International Journal of Engineering and Manufacturing (IJEM), Vol.16, No.5, pp. 384-410, 2026. DOI:10.5815/ijem.2026.05.21

Reference

[1]Y. Zhang, Z. Yin, Y. Li, G. Yin, J. Yan, J. Shao, and Z. Liu, "CelebA-Spoof: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations," in Proc. Eur. Conf. Comput. Vis. (ECCV), Cham, Switzerland: Springer, 2020, pp. 70–85, doi: 10.1007/978-3-030-58610-2_5.
[2]S. Zhang, A. Liu, J. Wan, Y. Liang, G. Guo, S. Escalera, H. J. Escalante, and S. Z. Li, "CASIA-SURF: A Large-Scale Multi-Modal Benchmark for Face Anti-Spoofing," IEEE Trans. Biometrics, Behav., Identity Sci., vol. 2, no. 2, pp. 182–193, Apr. 2020, doi: 10.1109/TBIOM.2020.2973001.
[3]A. Liu, Z. Tan, X. Li, J. Wan, S. Escalera, G. Guo, and S. Z. Li, "CASIA-SURF CeFA: A Benchmark for Multi-Modal Cross-Ethnicity Face Anti-Spoofing," in Proc. IEEE Winter Conf. Appl. Comput. Vis. (WACV), 2021, 1178–1186, doi: 10.1109/WACV48630.2021.00122.
[4]A. Alkholid, D. Islamaj, E. Qershori, and H. Hamam, "Facial Spoof Detection Using Deep Learning Techniques for Enhanced Biometric Security," in AI and Digital Transformation: Opportunities, Challenges, and Emerging Threats in Technology, Business, and Security, K. Dhoska and E. Spaho, Eds., Communications in Computer and Information Science (CCIS), vol. 2669, Cham, Switzerland: Springer, 2026, 46–58, doi: 10.1007/978-3-032-07373-0_3.
[5]D. Belli, D. Das, B. Major, and F. Porikli, "A Personalized Benchmark for Face Anti-Spoofing," in Proc. IEEE/CVF Winter Conf. Appl. Comput. Vis. Workshops (WACVW), 2022, pp. 338–348, doi: 10.1109/WACVW54805.2022.00040.
[6]R. Shao, X. Lan, and P. C. Yuen, "Joint Discriminative Learning of Deep Dynamic Textures for 3D Mask Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 14, no. 4, pp. 923–938, Apr. 2019, doi: 10.1109/TIFS.2018.2868230.
[7]H. Chen, G. Hu, Z. Lei, Y. Chen, N. Robertson, and S. Z. Li, "Attention-Based Two-Stream Convolutional Networks for Face Spoofing Detection," IEEE Trans. Inf. Forensics Security, vol. 15, pp. 578–593, 2020, doi: 10.1109/TIFS.2019.2922241
[8]D. Wen, H. Han, and A. K. Jain, "Face Spoof Detection With Image Distortion Analysis," IEEE Trans. Inf. Forensics Security, vol. 10, no. 4, pp. 746–761, Apr. 2015, doi: 10.1109/TIFS.2015.2400395.
[9]A. Pinto, H. Pedrini, W. R. Schwartz, and A. Rocha, "Face Spoofing Detection Through Visual Codebooks of Spectral Temporal Cubes," IEEE Trans. Image Process., vol. 24, no. 12, pp. 4726–4740, Dec. 2015, doi: 10.1109/TIP.2015.2466088.
[10]S. Lim, S. Kim, H. Kim, and Y. Ro, "One-Class Learning Method Based on Live Correlation Loss for Face Anti-Spoofing," IEEE Access, vol. 8, pp. 201635–201648, 2020, doi: 10.1109/ACCESS.2020.3035747.
[11]A. Liu et al., "Contrastive Context-Aware Learning for 3D High-Fidelity Mask Face Presentation Attack Detection," IEEE Trans. Inf. Forensics Security, vol. 17, pp. 2497–2507, 2022, doi: 10.1109/TIFS.2022.3188149.
[12]H. Fang, A. Liu, J. Wan, S. Escalera, C. Zhao, X. Zhang, S. Z. Li, and Z. Lei, "Surveillance Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 19, pp. 1535–1546, 2024, doi: 10.1109/TIFS.2023.3337970.
[13]A. Liu, S. Xue, J. Gan, J. Wan, Y. Liang, J. Deng, S. Escalera, and Z. Lei, "CFPL-FAS: Class-Free Prompt Learning for Generalizable Face Anti-Spoofing," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Seattle, WA, USA, 2024, 222–232, doi: 10.1109/CVPR52733.2024.00014.
[14]H. Fang et al., "Unified Physical-Digital Face Attack Detection," in Proc. 33rd Int. Joint Conf. Artif. Intell. (IJCAI), 2024, 749–757, doi: 10.24963/ijcai.2024/93.
[15]Z. Li, Y. Li, J. Wan, and Z. Lei, "Optimal Transport-Guided Source-Free Adaptation for Face Anti-Spoofing," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), 2025, pp. 24351–24363, doi: 10.1109/CVPR52734.2025.02268.
[16]Y. Liu, X. Zhang, J. Wan, and Z. Lei, "Dual Consistency Regularization for Generalized Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 20, pp. 2171–2183, 2025, doi: 10.1109/TIFS.2025.3540659.
[17]C.-H. Liao, W.-C. Chen, H.-T. Liu, Y.-R. Yeh, M.-C. Hu, and C.-S. Chen, "Domain Invariant Vision Transformer Learning for Face Anti-Spoofing," in Proc. IEEE/CVF Winter Conf. Appl. Comput. Vis. (WACV), 2023, 6087–6096, doi: 10.1109/WACV56688.2023.00605.
[18]Z. Wang, Q. Wang, W. Deng, and G. Guo, "Learning Multi-Granularity Temporal Characteristics for Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 17, pp. 1254–1269, 2022, doi: 10.1109/TIFS.2022.3158062.
[19]W. Zhang, H. Liu, F. Liu, R. Ramachandra, and C. Busch, "Effective Presentation Attack Detection Driven by Face-Related Tasks," in Proc. Eur. Conf. Comput. Vis. (ECCV), Cham, Switzerland: Springer, 2022, pp. 408–423, doi: 10.1007/978-3-031-20065-6_24.
[20]M. A. Sánchez-Sánchez, C. Conde, B. Gómez-Ayllón, D. Ortega-DelCampo, A. Tsitiridis, D. Palacios-Alonso, and E. Cabello, "Convolutional Neural Network Approach for Multispectral Facial Presentation Attack Detection in Automated Border Control Systems," Entropy, vol. 22, no. 11, Art. no. 1296, Nov. 2020, doi: 10.3390/e22111296.
[21]X. Long, J. Zhang, S. Wu, X. Jin, and S. Shan, "Dual Sampling Based Causal Intervention for Face Anti-Spoofing With Identity Debiasing," IEEE Trans. Inf. Forensics Security, vol. 19, pp. 851–862, 2024, doi: 10.1109/TIFS.2023.3326370
[22]Y. Jia, J. Zhang, S. Shan, and X. Chen, "Dual-Branch Meta-Learning Network With Distribution Alignment for Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 17, pp. 138–151, 2022, doi: 10.1109/TIFS.2021.3134869.
[23]Y. Jia, J. Zhang, S. Shan, and X. Chen, "Unified Unsupervised and Semi-Supervised Domain Adaptation Network for Cross-Scenario Face Anti-Spoofing," Pattern Recognit., vol. 115, Art. no. 107888, Jul. 2021, doi: 10.1016/j.patcog.2021.107888.
[24]X. Long and J. Zhang, "Confidence Aware Learning for Reliable Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 20, pp. 5083–5093, 2025, doi: 10.1109/TIFS.2025.3564878
[25]X. Long, J. Zhang, and S. Shan, "Generalized Face Liveness Detection via De-fake Face Generator," IEEE Trans. Pattern Anal. Mach. Intell., vol. 47, no. 3, pp. 1818–1831, Mar. 2025, doi: 10.1109/TPAMI.2024.3507101
[26]Y. Zhang, Y. Liu, J. Wan, and Z. Lei, "Robust Face Anti-Spoofing With Dual Probabilistic Modeling," Pattern Recognit., vol. 167, Art. no. 111700, Nov. 2025, doi: 10.1016/j.patcog.2025.111700.
[27]M. Feng, Y. Ito, and T. Okatani, "Leveraging Intermediate Features of Vision Transformer for Face Anti-Spoofing," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Workshops (CVPRW), 2025, 3464–3472, doi: 10.1109/CVPRW63200.2025.00095.
[28]Y. Tian, K. Zhang, L. Wang, and Z. Sun, "Cross-Optical Property Image Translation for Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 19, pp. 6092–6104, 2024, doi: 10.1109/TIFS.2024.3521323
[29]R. Cai, Z. Li, R. Wan, H. Li, Y. Hu, and A. C. Kot, "Learning Meta Pattern for Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 17, pp. 1201–1213, 2022, doi: 10.1109/TIFS.2022.3158551.
[30]R. Cai, H. Li, S. Wang, C. Chen, and A. C. Kot, "DRL-FAS: A Novel Framework Based on Deep Reinforcement Learning for Face Anti-Spoofing," IEEE Trans. Inf. Forensics Security, vol. 16, pp. 937–951, 2021, doi: 10.1109/TIFS.2020.3026553
[31]M. Radad, A. A. Aburomman, and M. Alkasassbeh, "Face Anti-Spoofing Detection Based on Novel Encoder Convolutional Neural Network and Texture Features," Int. J. Inf. Secur., vol. 24, no. 1, Art. no. 45, 2025, doi: 10.1007/s44196-025-00757-z
[32]C. Kong, K. Zheng, Y. Liu, S. Wang, A. Rocha, and H. Li, "M³FAS: An Accurate and Robust MultiModal Mobile Face Anti-Spoofing System," IEEE Trans. Dependable Secure Comput., vol. 21, no. 6, pp. 5650–5666, Nov./Dec. 2024, doi: 10.1109/TDSC.2024.3381598.
[33]C. Kong, K. Zheng, S. Wang, A. Rocha, and H. Li, "Beyond the Pixel World: A Novel Acoustic-Based Face Anti-Spoofing System for Smartphones," IEEE Trans. Inf. Forensics Security, vol. 17, pp. 3238–3253, 2022, doi: 10.1109/TIFS.2022.3202115
[34]R. Cai, S. Wang, H. Li, Y. Hu, X. Liu, and A. C. Kot, "S-Adapter: Generalizing Vision Transformer for Face Anti-Spoofing With Statistical Tokens," IEEE Trans. Inf. Forensics Security, vol. 19, pp. 8385–8397, 2024, doi: 10.1109/TIFS.2024.3420699.
[35]Z. Yu, R. Cai, Y. Cui, X. Liu, Y. Hu, and A. C. Kot, "Rethinking Vision Transformer and Masked Autoencoder in Multimodal Face Anti-Spoofing," Int. J. Comput. Vis., vol. 132, 11, pp. 5217–5238, Dec. 2024, doi: 10.1007/s11263-024-02055-1
[36]X. Lin, S. Wang, R. Cai, Y. Liu, Y. Fu, W. Tang, Z. Yu, and A. C. Kot, "Suppress and Rebalance: Towards Generalized Multi-Modal Face Anti-Spoofing," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Seattle, WA, USA, 2024, pp. 211–221, doi: 10.1109/CVPR52733.2024.00028.
[37]A. Liu, Z. Tan, J. Wan, Y. Liang, Z. Lei, G. Guo, and S. Z. Li, "Face Anti-Spoofing via Adversarial Cross-Modality Translation," IEEE Trans. Inf. Forensics Security, vol. 16, pp. 2759–2772, 2021, doi: 10.1109/TIFS.2021.3065495
[38]H. Xue, Y. Zhang, Z. Wang, and X. Li, "A Hierarchical Multi-Modal Cross-Attention Model for Face Anti-Spoofing," J. Vis. Commun. Image Represent., 97, Art. no. 103969, Jul. 2023, doi: 10.1016/j.jvcir.2023.103969.
[39]J. X. Chong, Y. Hsu, M. T. Hsu, Y. T. Lin, K. H. Chien, C. T. Hsu, and P. K. Huang, "Multi-Modal Face Anti-Spoofing via Cross-Modal Feature Transitions," Expert Syst. Appl., vol. 310, Art. no. 131292, May 2026, doi: 10.1016/j.eswa.2026.131292.
[40]Y. Yan, L. Shi, Y. Zhang, and X. Chang, "Multimodal feature enhancement via dynamically-aware heterogeneous network for face anti-spoofing," Sci. Rep., vol. 16, Art. no. 3535, Dec. 2025, doi: 10.1038/s41598-025-33502-9.
[41]N. Li, Y. Yan, X. Zhang, and H. Wang, "Dynamic Cross-Modal Modeling With an Ultra-Lightweight Framework for Face Anti-Spoofing," Sensors, vol. 26, no. 9, Art. no. 2699, May 2026, doi: 10.3390/s26092699.
[42]J. P. Appadurai, R. S. N. Noella, M. S. Jacob, K. Arunasakthi, B. S. K. Devi, and M. K. Singh, "An Effective Biometric Medical Image Watermarking System Designed for e-Health Application," in Machine Learning Algorithms for Data Security and Healthcare Applications, Communications in Computer and Information Science (CCIS), Cham, Switzerland: Springer, 2025, 310–317, doi: 10.1007/978-3-031-75861-4_27.
[43]M. K. Singh, S. Kumar, R. Ranjan, and D. Nandan, "Duplicate Image Detection Using Deep Learning Modified SVM and k-NN Classification Method for Multimedia Application," Soft Comput., vol. 28, no. 13–14, pp. 7659–7670, Jul. 2024, doi: 10.1007/s00500-024-09756-2.
[44]M. Suneel and T. Ranga Babu, "Multi-modal Face Anti-Spoofing Identification Using Efficient Networks Integrated with U-Adapter for Unreliable Region Mitigation," SSRG Int. J. Electron. Commun. Eng., vol. 12, no. 11, pp. 133–145, Nov. 2025, doi: 10.14445/23488549/IJECE-V12I11P111.
[45]S. Mudunuru and T. R. Babu, "Efficient Face Anti-Spoofing Identifier Network (FASIN) with Depth and Near Infrared Deep Learning Methods," ARPN J. Eng. Appl. Sci., vol. 19, no. 20, pp. 1255–1271, Oct. 2024, doi: 10.59018/102457.
[46]Z. Mostaani, A. George, G. Heusch, D. Geissbühler, and S. Marcel, "The High-Quality Wide Multi-Channel Attack (HQ-WMCA) Database," Idiap Research Report Idiap-RR-22-2020, Idiap Research Institute, 2020.
[47]Y. Liu, A. Jourabloo, and X. Liu, "Deep Tree Learning for Zero-Shot Face Anti-Spoofing," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), 2019, 4675–4684, doi: 10.1109/CVPR.2019.00481.
[48]A. Pinto, H. Pedrini, W. R. Schwartz, A. Rocha, "Mobile face spoofing challenge," in Proc. IEEE 7th Int. Conf. Biometrics: Theory, Applications and Systems (BTAS), 2015.

International Journal of Engineering and Manufacturing (IJEM)