Muhammad Naseer Subhani

I’m an Industrial AI & Embedded Vision Engineer specializing in building scalable, production-grade AI systems that bridge cutting-edge research with real-world industrial deployment. My work focuses on improving efficiency, reliability, and automation in manufacturing and embedded environments.

I design and deploy end-to-end AI solutions—from model development to optimized inference on constrained hardware—driving measurable impact in production systems.

• Core areas: industrial inspection, anomaly detection, detection/segmentation, domain adaptation
• Edge & deployment: NVIDIA Jetson (Xavier NX, TX2, AGX), low-power systems, real-time pipelines
• AI systems: CV pipelines, AI agents, LLM fine-tuning (LoRA), automation workflows
• Tooling: PyTorch, TensorFlow/Keras, OpenCV, ROS2, Docker, CI/CD

News

• 2026-02: Accepted at CVPR 2026
  Paper (arXiv PDF) · Code
• 2020-07: Paper accepted at ECCV 2020

Publications

• CVPR 2026 — ReSAM: Refine, Requery, and Reinforce: Self-Prompting Point-Supervised Segmentation for Remote Sensing Images
  Paper (arXiv PDF) · Code

• ECCV 2020 — Learning from Scale-Invariant Examples for Domain Adaptation in Semantic Segmentation
  Paper (arXiv PDF)

Honors & Awards

• First solo author from Pakistan at CVPR
• Awarded $10,000 Azure credits by Microsoft
• Gold Medal — Microcontroller Programming (UCP Olympic 2015)
• Silver Medal — Microcontroller Programming (UCP Olympic 2016)
• 1st Position — Microcontroller Programming (UOL ECL 2016)

Contact

• Email: im.naseerr@gmail.com