About Me

I’m a Computer Vision Engineer with a strong foundation in mathematics and research, currently working in the Avatar Team at Colossyan. My work bridges the gap between cutting-edge machine learning research and scalable, real-world applications.

I specialize in:

• 3D Reconstruction
• Dynamic Gaussian Splatting
• Neural Radiance Fields (NeRF)
• Motion Blur Reduction
• Generative AI and Avatar Systems

🔬 Research & Collaboration

At the Intelligent Machines Lab (ITU Lahore), I’ve collaborated with ETRI South Korea to tackle motion blur in dynamic 3D scenes and contribute to smoother mesh generation from volumetric NeRF data. My research is published in high-impact venues including:

• IEEE Access
• Multimedia Tools and Applications

🧠 Technical Background

I have experience building and deploying machine learning systems end-to-end, with a focus on research-grade performance and production-level robustness. My core technical skills include:

• Programming: Python, PyTorch, TorchScript, OpenCV, C++, CUDA, STL
• Technologies: Deep Learning, Computer Vision, Multi-View Geometry, 3D Vision
• DevOps & Cloud: AWS (S3, Lambda, SageMaker), Docker, GitHub Actions, ArgoWorkflows
• Other Tools: Scikit-learn, Flask, Matplotlib

💼 Experience

• Machine Learning Engineer @ Colossyan (2024 – Present)
  Working on the Avatar team to improve generation pipelines and backend tooling.

• Research Intern / Associate @ Intelligent Machines Lab, ITU (2022 – Present)
  Developed NeRF-based systems, worked on motion blur reduction in 3D Gaussian Splatting.

• Computer Vision Engineer @ DeepGears (2021)
  Created 3D pose estimation and mesh reconstruction systems for AR applications.

• Research Assistant @ NCAI, PUCIT (2020)
  Worked on LiDAR-based object detection for autonomous vehicles.

📂 Open-Source Projects

• Codebase for R-CNN:
  An object detection framework inspired by Detectron2 and MMDetection.

• AEPI: Ear Recognition:
  A biometric recognition project using deep learning on ear images.

I’m always looking to connect with researchers, engineers, and innovators. Feel free to reach out or check out my GitHub and LinkedIn for more.

Contact me

usamahasan72@gmail.com

Selected Publications.

AEPI: insights into the potential of deep representations for human identification through outer ear images U Hasan, W Hussain, N Rasool Multimedia Tools and Applications 81 (8), 10427-10443

CloudUP - Upsampling Vibrant Color Point Clouds using Multi-Scale Spatial Attention (Accepted IEEE ACESS)