A deep dive into optimizing a full-stack perception engine using YOLOv8 and SORT, achieving 127 FPS through TensorRT and FP16 quantization.

This post explores road segmentation using a U-Net architecture trained on monocular images from the KITTI dataset. The task focuses on identifying drivable areas from RGB inputs, a challenge due to lighting, shadows, and variable road conditions. I detail the model architecture, training strategies, and key hyperparameters such as learning rate selection and loss functions.

A SORT-based object tracking framework for monocular camera sensors in an automotive environment. Tested against KITTI benchmark.

Design and implementation of Extended and Unscented Kalman Filters in Python using numpy and scipy. Experiment to evaluate their performance and scope.

A CNN that classifies FMCW radar targets based on the surrounding cells of the range-Doppler map. Several network architectures were tested with special attention to model size and number of parameters to provide a lightweight solution.

A dashboard-like website that provides statistics about the world COVID-19 pandemic situation. With daily updates, the website allows quick access to your country's data using IP recognition as well as a comparison between multiple countries.