Md Al Amin
Computer vision researcher working on fine-grained visual classification, explainable AI and biometric security.
I hold an M.Sc. in Computer Engineering from Chosun University, South Korea, where I worked as a Research Assistant in the Computer Vision Lab under Professor Ho Yub Jung on a fully funded scholarship. My work focuses on localized and receptive field-wise feature learning for fine-grained image recognition and lightweight networks for spoof fingerprint detection. Before that, I worked on bioinformatics and machine learning for healthcare. I am now looking to pursue a PhD in explainable AI and cybersecurity.
Research
Main areas I work in and recent highlights.
MCL: Meta Convolution Layer
A meta convolution layer built around a meta-block, proposed as a new building block for convolutional neural networks.
arXiv preprint, 2026 Read paper → SurveyFine-Grained Visual Classification: A Survey of Methods, Architectures, and Emerging Trends
A survey of methods, architectures and emerging trends in fine-grained visual classification.
IEEE Access, 2026 Read paper → Fine-grained recognitionLocalized Feature Enhancement via Patch-Wise Learning for Fine-Grained Image Classification
Patch-wise learning to enhance localized features for fine-grained image classification.
IEEE Access, 2025 Read paper → Biometric securityLightweight Network for Spoof Fingerprint Detection by Attention-Aggregated Receptive Field-Wise Feature
A lightweight network with only 0.14M parameters for spoof fingerprint detection, evaluated on LivDet-2015 and LivDet-2017.
Electronics, 2025 Read paper →Experience
Research Assistant, Computer Vision Lab
Chosun University, South Korea · Advisor: Prof. Ho Yub Jung
- Research on localized receptive field-wise image feature learning (M.Sc. thesis).
- First-authored work on fine-grained image classification and spoof fingerprint detection in IEEE Access and Electronics.
- Co-developed the Meta Convolution Layer (MCL), available as an arXiv preprint.
Undergraduate and Collaborative Research
Prime University, Bangladesh and collaborators
- Statistical and network-based bioinformatics studies on type 2 diabetes, COVID-19 and related diseases.
- Machine learning for healthcare, including diabetes prediction and IoMT-based remote monitoring.
Teaching Experience
Alongside my own research, I have taught and guided junior students in research. Here are some of them, in their own words.
Publications
Citation counts from Google Scholar, October 2026.
Education
M.Sc. in Computer Engineering
Chosun University, South Korea
Thesis: Localized Receptive Field-wise Image Feature Learning
CGPA: 4.06 / 4.50 · Fully funded scholarship
B.Sc. in Computer Science and Engineering
Prime University, Bangladesh
Thesis: Using Statistical Bioinformatics to Reveal the Biological Mechanisms Linking Smoking and Type 2 Diabetes through Transcriptomic and GWAS Analysis
CGPA: 3.81 / 4.00
Projects
- Computer VisionFine-grained image classification with CNNs in PyTorch
- Computer VisionFake fingerprint detection with CNNs in PyTorch
- Computer VisionSteering angle prediction from lane line images
- Generative ModelsDigit image generation with GANs on MNIST
- NLP20 Newsgroups classification with attention-based BiLSTM
- NLPQuora question pair similarity
Awards and Certifications
- 2023Fully funded scholarship for M.Sc. in Computer Engineering, Chosun University
- 2020Runner-up, Programming Contest, Dept. of CSE, Prime University
- CertificationPython Basic and Intermediate, HackerRank
- CertificationCOVID-19 Operational Planning Guidelines and Partners Platform, WHO
- Fellowship4th bdSIG, Bangladesh School of Internet Governance
Let's work together
I am open to PhD positions, research collaborations and conversations about computer vision and explainable AI.