Computer Vision
Image classification, segmentation, remote sensing, and Vision Transformer workflows.
NSF GRFP Fellow + Computer Vision Researcher
Talha Mahmood
Computer Science PhD student at the University of Delaware building multimodal deep learning systems for multispectral, geospatial, and scientific datasets.
- NSF
- Graduate Research Fellow
- PhD
- Computer Science at UD
- CASE
- Research Assistant
// about
PhD researcher turning computer vision into scientific tools.
I am an NSF Graduate Research Fellow and Computer Science PhD student in the Department of Computer & Information Sciences at the University of Delaware, advised by Dr. Yuna. My research at the CASE Lab focuses on image classification and multimodal deep learning with multispectral and geospatial datasets.
I specialize in architecting robust deep learning models for real-world scientific challenges, translating advances in computer vision into meaningful AI applications for researchers, educators, and the broader scientific community.
$ whoami
talha_mahmood
$ focus
multispectral_data, geospatial_ai, multimodal_learning
$ current
CASE_Lab research assistant advised by Dr. YunaMultimodal AI
Deep learning pipelines for multispectral, hyperspectral, and geospatial observations.
Teaching
Graduate and undergraduate TA work across ML, data mining, logic, robotics, and CS foundations.
// updates
Recent Signals
- Selected as an NSF Graduate Research Fellowship Program Fellow.
- Published J-STARS work on benthic habitat trends in Guam using WorldView-2/3 observations and deep learning.
- Started the Computer Science PhD program at the University of Delaware.
- Graduated summa cum laude with distinction from the University of Delaware.
- Defended undergraduate senior thesis at the University of Delaware.
- Received the Computer and Information Sciences Outstanding Senior Student Award.
- Presented MMST-ViT crop yield prediction research at the Symposium for Undergraduate Research and Creative Activity.
- Presented Medical Marauders hospital ranking research at the Data Science Institute Symposium after winning the Most Impactful Project Award.
- Won Best Educational Technology Hack at HenHacks with Team BreakOut.
// research
Deep learning for scientific observation.
My research centers on computer vision, multimodal AI, and machine learning for science, with an emphasis on robust models for multispectral, hyperspectral, and geospatial data.
J-STARS publication
Benthic Habitat Mapping from WorldView-2/3
Advanced deep learning for biogeographical trends of benthic habitats in Guam, using WorldView-2/3 observations to support marine ecosystem monitoring and conservation.
IEEE J-STARS
Spatial-temporal AI
Climate-aware Crop Yield Prediction
A Multi-Modal Spatial-Temporal Vision Transformer approach for soybean yield prediction using remote sensing and meteorological data, built to reason across environmental patterns over time.
Research page// publications
Peer-reviewed Work
Research at the intersection of computer vision, geospatial data, and environmental science.
May 6, 2024
Biogeographical Trends of Benthic Habitats in Guam from WorldView-2/3 Observations Using Advanced Deep Learning
Mahmood, T., Liu, B., & Yuan, X. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS).
2024-2023
Selected Presentations
Research talks and posters across crop yield prediction, coral reef segmentation, and interpretable healthcare ranking models.
// experience
Research & Teaching
Department of Computer & Information Sciences, University of Delaware.
May 2024 - Present
Full-time · Newark, Delaware
Research Assistant, CASE Lab
Developing deep learning systems for scientific datasets, including cross-sensor fusion to downscale NASA PACE hyperspectral data with Sentinel-2 and marine ecosystem models using WorldView and PlanetScope imagery.
Jun 2025 - May 2026
Graduate Teaching Assistant
Machine Learning, Data Mining, AI in Healthcare, Logic in Computer Science
Supported graduate and undergraduate learning through course staff work in machine learning, data mining, AI in healthcare, and logic-centered computer science foundations.
Aug 2022 - May 2025
Part-time · Hybrid
Undergraduate Teaching Assistant
Supported students across Data Structures, Automata Theory, Mobile Robot Programming, Introduction to Computer Science, and General Computer Science for Engineers.
// projects
Selected Builds
React + TypeScript + MongoDB
Project X Movie Lists
A role-based movie list platform where admins, super users, and regular users manage shared catalog data, personal watch lists, and ratings.
Open project
Machine learning + healthcare data
Medical Marauders
Most Impactful Project Award winner at UD's Data Science + AI Summer Hackathon, predicting U.S. hospital rankings from CMS Hospital Compare data with interpretable random forest models.
Open project
Best Educational Technology Hack
BreakOut
HenHacks-winning educational technology project from Team BreakOut, built to help younger students learn programming concepts through approachable activities.
Open project
React + Firebase + AI
YOUniversity
A collaborative learning prototype with adaptive subject flows, real-time student support, and AI-assisted question updates.
Open project// stack
Tools I Reach For
Comfortable moving between deep learning research, data analysis, and production-minded software workflows.
AI + Data
- PyTorch
- TensorFlow
- OpenCV
- Computer Vision
- Vision Transformers
- Remote Sensing
- Multimodal Learning
Languages
- Python
- Java
- C++
- C
- TypeScript
- JavaScript
Libraries, Tools + Methods
- NumPy
- Pandas
- Scikit-learn
- React
- Git/GitHub
- Cross-sensor Fusion
- Model Optimization
- Technical Writing
// cv
Academic Snapshot
PhD student in Computer Science at the University of Delaware, NSF Graduate Research Fellow, and summa cum laude UD Computer Science graduate.
Ph.D.
Computer Science, University of Delaware
August 2025 - Present. NSF Graduate Research Fellow. Current GPA: 3.93.
B.S.
Computer Science with Distinction
Graduated summa cum laude in May 2025 with a 3.98 GPA, a concentration in AI & Robotics, and a Mathematics minor.
Recognition
NSF GRFP, Outstanding Senior Award, and hackathon honors
Selected for the NSF GRFP in April 2026 and received the CIS Outstanding Senior Student Award in March 2025; earned hackathon honors at HenHacks and UD's Data Science + AI Summer Hackathon.
// contact
Let us build something useful.
I am open to research collaborations in computer vision, multimodal AI, geospatial learning, scientific machine learning, and education-focused computing projects.