Yi-Chen Zhang

This post explains how I deployed the TorchVision FCOS object detection model in TensorRT for real-time inference — achieving a ~20× speedup over the LibTorch baseline.

This post outlines how I optimized a Fully Convolutional Network (FCN) segmentation pipeline — from 0.5 FPS (~2000ms per frame) to 31 FPS in real-time.

I've been experimenting with YOLOv11 recently, and I was genuinely surprised by how easy it is to use — it delivers accurate results with minimal effort.

Over the past few weeks, I have been studying the Invariant Extended Kalman Filter (IEKF) — a variant of the EKF specifically designed for systems with inherent symmetries or invariances, such as Lie group symmetries.

When people think about YOLO model versions, the assumption is usually straightforward: newer is better. But that isn't always the case — especially when you're using pre-trained weights without any fine-tuning.

I am pleased to share that my paper titled "Road Geometry Estimation Using Vehicle Trails: A Linear Mixed Model Approach" has been published in the Journal of Intelligent Transportation Systems: Technology, Planning, and Operations.