Abstract
Large-scale face feature matching is a critical issue for many practical applications where the computing process requires a burden of hardware resources and computational time. In this work, the so-called multi-probe locality sensitive hashing (LSH) is exploited to solve the large-scale face feature matching problem. The standard procedure for hashing the dataset into smaller buckets and a matching algorithm with some improvements were implemented. Experimental implementations on a dataset of 12,720,006 images, corresponding to 617,970 person identifications, were carried out. The results show that extended LSH outperforms excellently in terms of accuracy and computational time. In comparison to linear matching, the computing time is reduced from 12446 ms to 267 ms for data type FP32 bits, and from 5,794 ms to 192 ms for data type INT8, while the accuracy is reduced very slightly. The results show the high potential use of the algorithm for practical applications.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Wang, M., Deng, W.: Deep face recognition: a survey. arXiv:1804.06655 (2018)
Melikoglu, O., Ergin, O., Salami, B., Pavon, J., Unsal, O., Cristal, A.: A novel FPGA-based high throughput accelerator for binary search trees. arXiv:1912.01556 (2019)
Chen, H., Madaminov, Sergey., Ferdman, M., Milder, P.: FPGA-accelerated samplesort for large data sets. In: 2020 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA 2020), pp. 222–232. Association for Computing Machinery, USA (2020)
Cosine Similarity Using Xilinx Alveo. https://developer.xilinx.com/en/articles/cosine-similarity-using-xilinx-alveo.html. Accessed 20 Nov 2020
Dong, H.-N., Ha, N.-X., Tuan, D.-M.: A new approach for large-scale face-feature matching based on LSH and FPGA for edge processing. In: Tran, D.-T., Jeon, G., Nguyen, T.D.L., Lu, J., Xuan, T.-D. (eds.) ICISN 2021. LNNS, vol. 243, pp. 337–344. Springer, Singapore (2021). https://doi.org/10.1007/978-981-16-2094-2_42
Locality Sensitive Hashing. https://www.mit.edu/~andoni/LSH/. Accessed 20 Nov 2020
FALCONN Github Repository. https://github.com/FALCONN-LIB/FALCONN. Accessed 10 Oct 2021
Cormode, G., Dasgupta, A., Goyal, A., Lee, C.H.: An evaluation of multi-probe locality sensitive hashing for computing similarities over web-scale query logs. PLoS ONE 13(1) (2018). https://doi.org/10.1371/journal.pone.0191175
Lv, Q., Josephson, W., Wang, Z., Charikar, M., Li, K.: Multi-probe LSH: efficient indexing for high-dimensional similarity search. In: Event 33rd International Conference on Very Large Data Bases, VLDB 2007, Vienna, Austria, pp. 950–961. Association for Computing Machinery, Inc., Austria (2007)
Loi, T.L., Heo, J., Lee, J., Yoon, S.: VLSH: Voronoi-based locality sensitive hashing. In: IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, Tokyo (2013). https://doi.org/10.1109/IROS.2013.6697130
Andoni, A., Indyk, P., Laarhoven, T., Schmidt, L.: Practical and optimal LSH for angular distance. arXiv:1509.02897 (2015)
Moran, S., Lavrenko, V., Osborne, M.: Variable: bit quantisation for LSH. In: Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, vol. 2 Short Papers, Sofia, Bulgaria, pp. 753–758 (2013)
Eigen Homepage. https://eigen.tuxfamily.org/index.php?title=Main_Page. Accessed 18 Aug 2021
MegaFace Homepage. http://megaface.cs.washington.edu/. Accessed 18 Aug 2021
Acknowledgment
This work was funded by the CMC Institute of Science and Technology, CMC Corporation, Hanoi, Vietnam.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Nguyen-Xuan, H., Hoang-Nhu, D., Dang-Minh, T. (2022). An Evaluation of the Multi-probe Locality Sensitive Hashing for Large-Scale Face Feature Matching. In: Anh, N.L., Koh, SJ., Nguyen, T.D.L., Lloret, J., Nguyen, T.T. (eds) Intelligent Systems and Networks. Lecture Notes in Networks and Systems, vol 471. Springer, Singapore. https://doi.org/10.1007/978-981-19-3394-3_51
Download citation
DOI: https://doi.org/10.1007/978-981-19-3394-3_51
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-3393-6
Online ISBN: 978-981-19-3394-3
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)