Vision Field Notes

About

Vision Field Notes is a practical blog about computer vision and edge AI: training detectors, getting them onto small devices, and running them on live camera feeds without drowning in false alarms.

Who writes it

I am Maidul Hasan Masum, a computer vision engineer based in Malaysia. I have more than three years of production experience building and running video analytics on live camera feeds. In total that is more than 300 streams across workplace safety, traffic, retail and waste management. My production stack includes PyTorch, OpenCV, CUDA, TensorRT, ONNX, Docker and Triton. I am also finishing an MSc by research on efficient object detection for edge devices.

What you will find here

How I write

I avoid quoting numbers I cannot back up. If something depends on your hardware or footage, I say so and show you how to measure it. If you find a mistake, please tell me and I will correct it.

My own projects, including a browser tool for testing detection models, are on my portfolio and GitHub.