Seminar by Prof. Khoa Luu on 7/1/2020: Biometrics: From Theory to Practice

Time: 15:00 on 7/1/2020

Place: room 803, B1 building, HUST

Presenter:  Khoa LuuVinAI Senior Research Scientist & Assistant Professor, Director, Computer Vision and Image Understanding, University of Arkansas, USA

Title: Biometrics: From Theory to Practice


Biometrics is the branch of machine learning and computer vision that works on understanding visual data for body measurements and calculations. It refers to metrics related to human characteristics for authentication and identification. Over the last few years there have been many important breakthroughs in the theory and practice of biometrics. Among these practices, deep learning has emerged as field leading the effort towards building the state-of-the-art systems. The practice of deep learning however still requires extensive experience and expertise to produce satisfactory results. Our experience in face processing and building state-of-the-art face detectors, facial landmarking, face recognition and 3D face modeling systems allows us to have a deep understanding of the field of deep learning. We have demonstrated ideas that have resulted in systems that have greatly outperformed competing groups in the very challenging and competitive field of face processing and real-world biometrics products.

In this talk, we will review the development of such sophisticated architectures in these related applications. We showcase state-of-the-art systems we have developed in applications of face preprocessing, face detection, face landmarking, face recognition, 3D face modeling and face video-analytics. Specifically, we showcase our state-of-the-art face detection and recognition systems, face analytics systems and present how to utilize the concepts of attacking these commercial face recognition systems. Finally, we will address the problem at hand – face recognition and reconstruction in real time and in low-cost devices.

Bio: Dr. Khoa Luu is a VinAI Senior Research Scientist and an Assistant Professor of Computer Science and Computer Engineering Department, University of Arkansas. He was the Research Project Director and Postdoc in Cylab Biometrics Center at Carnegie Mellon University (CMU). He has coauthored 70+ papers, some are in top-tier conferences (e.g. CVPR, ICCV) and journals (e.g. IJCV, TPAMI, TIP, etc.). Many of these papers are about biometrics, face recognition and detection, predicting future faces, temporal deep learning modeling, low-power deep learning and reinforcement learning. He has registered 4 patents and inventions and received 10+ awards and scholarships. He received two best paper awards in IEEE Intl. Conf. on Biometrics: Theory, Applications and Systems in 2011 and 2012. He was a vice chair of Montreal Chapter IEEE SMCS in Canada from September 2009 to March 2011. Dr. Luu has strong professional experience in both research and real-world deployed projects in million-scale biometrics, long-range face recognition, real-time face modeling, video understanding, and low-power deep learning technologies.


Related Publications

[1] C. N. Duong, K. Luu, K. G. Quach, N. Nguyen, E. Patterson, T. D. Bui and T. H. N. Le, “Automatic Face Aging in Videos via Deep Reinforcement Learning,” IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.

[2] C. Zhu, Y. Ran, K. Luu and M. Savvides, “Seeing Small Faces from Robust Anchor’s Perspective”, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018.

[3] C. N. Duong, K. G. Quach, K. Luu, T. H. N. Le, M. Savvides, “Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition,”, IEEE International Conference on Computer Vision (ICCV), 2017.

[4] C. Bhagavatula, C. Zhu, K. Luu and M. Savvides, “Faster Than Real-time Facial Alignment: A 3D Spatial Transformer Network Approach in Unconstrained Poses“, IEEE International Conference on Computer Vision (ICCV), 2017.

[5] C. N. Duong, K. Luu, K. G. Quach and T. D. Bui, “Deep Appearance Models: A Deep Boltzmann Machine Approach for Face Modeling,” International Journal of Computer Vision (IJCV), pages 1-19, 2018.

[6] K. Luu, M. Savvides, T.D.Bui and C.Y.Suen, “Compressed Submanifold Multifactor Analysis,” IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI), vol. 39, issue 3, pages 444-456, 2017.

[7] F. Xu, K. Luu and M. Savvides, “Spartans: Single-sample Periocular-based Alignment-robust Recognition Technique Applied to Non-frontal Scenarios,” IEEE Trans. on Image Processing (TIP), vol. 24, issue 12, pages 4780-4795, Dec. 2015.