Introduction

I am Zahid Hasan. I am currently working as a postdoctoral research associate at UMBC. I have completed my Ph.D. Student in the IS Department at UMBC under the supervision of Dr. Nirmalya Roy in MPSC laboratory.

Research Interest

Machine learning, Deep learning, Self-supervised representation learning, computer vision, Information theory, and their application to real-world problem-solving.

Current Research

Developing robust methods for image and video understanding using unlabeled raw data and contributing to computer science by proposing and validating Machine learning algorithms. Particularly, I am looking into developing a view-invariant vision system, and an information-preserving vision system from unlabeled data using self-supervised learning.

Achievement

Best paper awards SmartComp 2023
Best paper awards SmartComp 2022
- Video Presentation
Award: Best Ph.D. student research award link
- Video Presentation
Award: Our research has been recognized as the best IS, UMBC Ph.D. Student research link
- Video Presentation
Problem-solving Award: The IEEE SP cup 2015, Runner Up.
Award: NSF travel grant chase 2021
Dean’s List Award in Undergraduate

News

Presenting Tutorial at SmartComp: Link

Our team has won the best paper award in the SmartComp 2022. Well, I am also in the news too. twitter, MPSC lab. Thank you everyone for your kind wishes and contribution throughout this work.

I have been selected by my IS, umbc department to submit Microsoft Fellowship Proposal. I will be submitting my future research ideas to improve novel categories discovery via contrastive learning approaches.

I am mentoring undergraduate REU students to develop a federated learning scheme for remote PPG data.

Our recent paper got accepted at the IEEE/ACM Chase conference.

Publications

Published

Hasan, Z., Faridee, A. Z. M., Ahmed, M., Purushotham, S., Kwon, H., Lee, H., & Roy, N. (2023). Novel Categories Discovery from Probability Matrix Perspective. arXiv preprint arXiv:2307.03856.
Masud Ahmed, Zahid Hasan, Nirmalya Roy, et al. ‘An Online Continuous Semantic Segmentation Framework With Minimal Labeling Efforts’ in Proceedings of the 9th IEEE International Conference on Smart Computing (SmartComp), Nashville, Tennessee, USA, June 2023
Zahid Hasan, Masud Hasan, Abu Zaher Md Faridee, Sanjay Purushotham, Heesung Kwon, Hyungtae Lee and Nirmalya Roy. NEV-NCD: Negative Learning, Entropy, and Variance regularization based novel action categories discovery, in arXiv preprint, April 2023 (accepted for publication in ICIP-2023)
Maloy Kumar Devnath, Avijoy Chakma, Mohammad Saeid Anwar, Emon Dey, Zahid Hasan, Marc Conn, Biplab Pal, and Nirmalya Roy. A Systematic Study on Object Recognition Using Millimeter-wave Radar, in Proceedings of the 9th IEEE International Conference on Smart Computing (SmartComp), Nashville, Tennessee, USA, June 2023
Zahid Hasan, Abu Zaher Md Faridee, Masud Ahmed and Nirmalya Roy. SrPPG: Semi-Supervised Adversarial Learning for Remote Photoplethysmography with Noisy Data, in Proceedings of the 9th IEEE International Conference on Smart Computing, Nashville, Tennessee, USA, June 2023
Hasan, Z., Faridee, A. Z. M., Ahmed, M., & Roy, N. (2022, November). Self-rPPG: Learning the Optical & Physiological Mechanics of Remote Photoplethysmography with Self-Supervision. In 2022 IEEE/ACM CHASE (pp. 46-56). IEEE.
Z. Hasan, E. Dey, S. R. Ramamurthy, N. Roy, and A. Misra, “Rhythmedge: Enabling contactless heart rate estimation on the edge.” Accepted for publication in 2022 IEEE International Conference on Smart Computing (SMARTCOMP), 2022.
Z. Hasan, E. Dey, S. R. Ramamurthy, N. Roy, and A. Misra, “Demo: Rhythmedge: Enabling contactless heart rate estimation on the edge.” Accepted for publication in 2022 IEEE International Conference on Smart Computing (SMARTCOMP), 2022.
A. Z. M. Faridee, A. Chakma, Z. Hasan, N. Roy, and A. Misra, “CoDEm: Conditional Domain Embeddings for Scalable Human Activity Recognition” Accepted for publication in 2022 IEEE International Conference on Smart Computing (SMARTCOMP), 2022.
Masud Ahmed, Zahid Hasan, Naima Khan, Nirmalya Roy, Sanjay Purushotham, Aryya Gangopadhyay, and Suya You “Benchmarking domain adaptation for semantic segmentation”, Proc. SPIE 12124, Unmanned Systems Technology XXIV, 121240F (31 May 2022); https://doi.org/10.1117/12.2618548
Zahid Hasan and Mohammad Ariful Haque. “Robust Heart Rate Estimation from PPG Signal Utilizing The Harmonic Analysis Of Motion Signal” IEEE TENCON 2015.
Shikder Shafiul Bashar, A. H. M. Zadidul Karim, Md. Abdullah Al Mahmud and Zahid Hasan, “A Machine learning Approach for Heart Rate Estimation from PPG Signal using Random Forest” IEEE regional Conference 2019
Hasan, Zahid, and Nirmalya Roy. “Trending machine learning models in cyber‐physical building environment: A survey.” Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery (2021): e1422.
Miah, Md Sazal, Shikder Shafiul Bashar, AHM Zadidul Karim, and Zahid Hasan. “Heart Rate Extraction from Photoplethysmography Signal: A Multi Model Machine Learning Approach.” In 2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET), pp. 1-6. IEEE, 2020.
Hasan, Zahid, Sreenivasan Ramasamy Ramamurthy, and Nirmalya Roy. “CamSense: A camera-based contact-less heart activity monitoring.” Smart Health 23 (2022): 100240

Experiences

Lecturer Daffodil University (2015-2016) at Electrical and Electronic Engineering Responsibilities: Conducting undergraduate EEE classes, Labs, and Evaluation Conducted Courses: Digital Signal Processing (DSP), Communication Theory
Lecturer UAP (2016-2018) in Electrical and Electronic Engineering Responsibilities: Conducting undergraduate EEE classes, Labs, and Evaluation Conducted Courses: Random Signal Processing (RSP), Advanced DSP, Communication Theory, and Cellular Communication.
Internship in Philips (May 2019 - December 2019) Responsibilities: I conducted research under the supervision of Philips AI scientists to develop a robust clutter removal filter for heart ultrasound for better Doppler-based blood flow estimation. We propose a data-driven approach to avoid the computationally heavy SVD method for clutter removal and achieve high performing model to estimate blood flow from noisy heart ultrasound.
Research Assistant in UMBC (Spring 2020 - Ongoing)
Paper Review: Reviewed several papers on PerCom, DMKD, Smartcomp, IEEE workshop on ML for Big data, IoT journal, PMC Journal, IEEE Chase, IEEE-BSN

Old research

rPPG Model Development details
rPPG Edge Implementation details
rPPG dataset

Education

University of Maryland, Baltimore County
- Ph.D. student (2020 -) [CGPA 4.00/4.00]
Boston University
- Ph.D. Fellowship (2018-2020)
Bangladesh University of Engineering and Technology
- B.Sc in Electrical and Electronic Engineering (2010 -2015) [CGPA 3.88/4.00], Merit Position 23/195

Coding Skills

Deep Learning Libraries: Pytorch, Tensorflow, Keras
Programming Languages: MATLAB/Octave, Python, C, C++, R
Simulation Software: PSpice, Proteus, MATLAB Simulink

Hobbies

Active problem solver in pytorch forum. bio link
Shoot me a ping if you stay near me and want to hit for some 4-5 rated tennis match or 2000-2200 rated table tennis or 1800 rated chess.
- Table Tennis: Penhold grip, Beware of my backhand! Careful in the forehand.
- Tennis: Playing Style: Aggressive, a lot of unforced errors.
- Chess: Fan of g4 opening, I play E4 when it gets a bit serious.
Writing critical summary blogs of AI papers
- Deeply curious about underlying mathematics.
Blogging about the things I get stuck on. Hopefully, others do not.
Curious about research methods and academic writings.

Research Approach

I am interested in finding and solving research gaps.
I would spend most of my time to know the SoTA research while formulating ideas to address one of the research gaps.
Create a Hypothesis, State the underlying assumption, Create experimentation, Check if it aligns with the hypothesis, Back to creating the hypothesis.
- Nutshell version!

[//] https://workwithcarolyn.com/blog/digital-cv-guide