M. Usman Rafique

I am a researcher working on computer vision and machine learning. My research areas include image synthesis, image understanding, and scene parsing and segmentation. I have been working with natural, outdoor scenes and remote sensing images (both aerial and satellite).

Recently, I have been working on large language models (LLMs) such as GPT. I have implemented GPT-Nano, a light-weight alternative of GPT-2 and GPT-3. I am currently working on efficient fine-tuning a 20 billion GPT model on a single GPU (blog post coming soon).

During PhD from the University of Kentucky, I was a member of the Multimodal Vision Research Lab working with Dr. Nathan Jacobs. My co-advisor was Dr. Samson Cheung. My PhD research was about combining information from multiple images for scene understanding and image synthesis.

After finishing PhD, I worked at Kitware Inc. as a senior research and development engineer for 1 year and 9 months.

My old website from my teaching days is available here.

Update July 2023 I am now looking for new opportunities :) Feel free to reach out, my contact details are on the left panel.

Update Aug 2023 I have finished my job search; excited to join RnD Lab of Bastian Solutions, a Toyota Advanced Logistics Company, as a Senior Machine Learning Engineer. I will be doing research on computer vision models for robotics and automation.

Research Projects

Near-Remote Sensing
overview figure of near remote sensing project

Diverse View Synthesis

Novel View Synthesis

Multi-Image Fusion

Weakly Supervised Segmentation

Recent News