Abstract
With the commercial application of face recognition systems, face anti-spoofing has been studied extensively to enhance security in recent years. In this work, a lightweight network via knowledge distillation for face anti-spoofing is proposed. The main innovations of our approach are threefold: (1) In convolutional neural network based knowledge distillation, the local receptive field of teacher network may be inconsistent with that of student network, which results in misguiding. In our method, vision transformer architecture is leveraged because of its global modeling capabilities. (2) Beyond conventional decision-level knowledge transfer in the classification step via kullback-leibler loss, we present multi-stage feature-level knowledge distillation strategy to guide the feature learning of student network which can transfer richer knowledge from teacher to student network. (3) In contrast to traditional projection head learning, we construct a covariance matrix to solve the embedding dimensionality mismatching problem between teacher and student network in middle layers. Compared to teacher model of 1.28 GB, the memory of student model is only 330.8 MB, which effectively achieves a trade-off between memory and accuracy. Extensive experiments on three standard benchmarks demonstrate the superiority of our proposed method, which evidently corroborates the significance of multi-stage knowledge distillation for face anti-spoofing.
This work was supported by the National Natural Science Foundation of China (62101376) and Natural Science Foundation of Shanxi Province of China (201901D211078).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Boulkenafet, Z., Komulainen, J., Hadid, A.: Face anti-spoofing based on color texture analysis. In: IEEE international conference on image processing. pp. 2636–2640 (2015)
Boulkenafet, Z., Komulainen, J., Li, L., Feng, X., Hadid, A.: Oulu-npu: A mobile face presentation attack database with real-world variations. In: IEEE international conference on automatic face & gesture recognition. pp. 612–618 (2017)
Chen, D., Mei, J.P., Zhang, H., Wang, C., Feng, Y., Chen, C.: Knowledge distillation with the reused teacher classifier. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 11933–11942 (2022)
Chen, H., Hu, G., Lei, Z., Chen, Y., Robertson, N.M., Li, S.Z.: Attention-based two-stream convolutional networks for face spoofing detection. IEEE Trans. Inf. Forensics Secur. 15, 578–593 (2019)
Chingovska, I., Anjos, A., Marcel, S.: On the effectiveness of local binary patterns in face anti-spoofing. In: Proceedings of the international conference of biometrics special interest group. pp. 1–7 (2012)
Chingovska, I., Dos Anjos, A.R., Marcel, S.: Biometrics evaluation under spoofing attacks. IEEE Trans. Inf. Forensics Secur. 9(12), 2264–2276 (2014)
Choudhary, T., Mishra, V., Goswami, A., Sarangapani, J.: A comprehensive survey on model compression and acceleration. Artif. Intell. Rev. 53, 5113–5155 (2020)
Deng, Y.: Deep learning on mobile devices: a review. In: Mobile Multimedia/Image Processing, Security, and Applications 2019. vol. 10993, pp. 52–66 (2019)
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, et al.: An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learning Representations (2021)
Galbally, J., Marcel, S., Fierrez, J.: Biometric antispoofing methods: A survey in face recognition. IEEE Access 2, 1530–1552 (2014)
Geras, K.J., Mohamed, A.r., Caruana, R., Urban, G., Wang, S., Aslan, O., Philipose, M., Richardson, M., Sutton, C.: Blending lstms into cnns. In: International Conference on Learning Representations (2016)
Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network. In: NIPS Deep Learning and Representation Learning Workshop (2015)
Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., Adam, H.: Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861 (2017)
Huang, Y.H., Hsieh, J.W., Chang, M.C., Ke, L., Lyu, S., Santra, A.S.: Multi-teacher single-student visual transformer with multi-level attention for face spoofing detection. In: British Machine Vision Conference. p. 125 (2021)
Kim, J., Park, S., Kwak, N.: Paraphrasing complex network: Network compression via factor transfer. Advances in neural information processing systems 31 (2018)
Komulainen, J., Hadid, A., Pietikäinen, M.: Context based face anti-spoofing. In: IEEE Sixth International Conference on Biometrics: Theory, Applications and Systems. pp. 1–8 (2013)
Laurens, Van Der Maaten, and Hinton, Geoffrey: Visualizing data using t-sne. Journal of Machine Learning Research 9(2605), 2579–2605 (2008)
Li, H., He, P., Wang, S., Rocha, A., Jiang, X., Kot, A.C.: Learning generalized deep feature representation for face anti-spoofing. IEEE Trans. Inf. Forensics Secur. 13(10), 2639–2652 (2018)
Li, H., Wang, S., He, P., Rocha, A.: Face anti-spoofing with deep neural network distillation. IEEE Journal of Selected Topics in Signal Processing 14(5), 933–946 (2020)
Li, S., Dong, J., Chen, J., Gao, X., Niu, S.: Vision transformer with depth auxiliary information for face anti-spoofing. In: Neural Information Processing: 29th International Conference, Part III. pp. 335–346 (2023)
Li, Z., Cai, R., Li, H., Lam, K.Y., Hu, Y., Kot, A.C.: One-class knowledge distillation for face presentation attack detection. IEEE Trans. Inf. Forensics Secur. 17, 2137–2150 (2022)
Passban, P., Wu, Y., Rezagholizadeh, M., Liu, Q.: Alp-kd: Attention-based layer projection for knowledge distillation. In: Proceedings of the AAAI Conference on artificial intelligence. vol. 35, pp. 13657–13665 (2021)
Patel, K., Han, H., Jain, A.K.: Secure face unlock: Spoof detection on smartphones. IEEE Trans. Inf. Forensics Secur. 11(10), 2268–2283 (2016)
Peng, F., Meng, S.h., Long, M.: Presentation attack detection based on two-stream vision transformers with self-attention fusion. Journal of Visual Communication and Image Representation 85, 103518 (2022)
Polino, A., Pascanu, R., Alistarh, D.: Model compression via distillation and quantization. arXiv preprint arXiv:1802.05668 (2018)
Ratha, N.K., Connell, J.H., Bolle, R.M.: Enhancing security and privacy in biometrics-based authentication systems. IBM Syst. J. 40(3), 614–634 (2001)
Romero, A., Ballas, N., Kahou, S.E., Chassang, A., Bengio, Y.: Fitnets: Hints for thin deep nets. In: International Conference on Learning Representations (2015)
Touvron, H., Cord, M., Douze, M., Massa, F., Sablayrolles, A., Jégou, H.: Training data-efficient image transformers & distillation through attention. In: International conference on machine learning. pp. 10347–10357 (2021)
Yang, J., Lei, Z., Li, S.Z.: Learn convolutional neural network for face anti-spoofing. arXiv preprint arXiv:1408.5601 (2014)
Yang, X., Luo, W., Bao, L., Gao, Y., Gong, D., Zheng, S., Li, Z., Liu, W.: Face anti-spoofing: Model matters, so does data. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 3507–3516 (2019)
Yang, Z., Li, Z., Zeng, A., Li, Z., Yuan, C., Li, Y.: Vitkd: Practical guidelines for vit feature knowledge distillation. arXiv preprint arXiv:2209.02432 (2022)
Yu, Z., Qin, Y., Li, X., Zhao, C., Lei, Z., Zhao, G.: Deep learning for face anti-spoofing: A survey. IEEE Trans. Pattern Anal. Mach. Intell. 45(5), 5609–5631 (2022)
Yu, Z., Zhao, C., Wang, Z., Qin, Y., Su, Z., Li, X., Zhou, F., Zhao, G.: Searching central difference convolutional networks for face anti-spoofing. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 5295–5305 (2020)
Zhang, L.B., Peng, F., Qin, L., Long, M.: Face spoofing detection based on color texture markov feature and support vector machine recursive feature elimination. J. Vis. Commun. Image Represent. 51, 56–69 (2018)
Zhang, Z., Yan, J., Liu, S., Lei, Z., Yi, D., Li, S.Z.: A face anti-spoofing database with diverse attacks. In: IAPR international conference on Biometrics. pp. 26–31 (2012)
Zhu, M., Tang, Y., Han, K.: Vision transformer pruning. arXiv preprint arXiv:2104.08500 (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhang, J., Zhang, Y., Shao, F., Ma, X., Zhou, D. (2024). KDFAS: Multi-stage Knowledge Distillation Vision Transformer for Face Anti-spoofing. In: Liu, Q., et al. Pattern Recognition and Computer Vision. PRCV 2023. Lecture Notes in Computer Science, vol 14429. Springer, Singapore. https://doi.org/10.1007/978-981-99-8469-5_13
Download citation
DOI: https://doi.org/10.1007/978-981-99-8469-5_13
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-8468-8
Online ISBN: 978-981-99-8469-5
eBook Packages: Computer ScienceComputer Science (R0)