The goal of CarZam was to create a highly accurate tool to predict the make and model of a vehicle. CarZam leverages state of the art technologies developed in summer of 2020.
CarZam is a vehicle make model recognition (VMMR) tool built with computer vision, object dectection and classification. Currently CarZam has consistently around 99% accuracy. CarZam utilizes the YOLO V5 framework, which is a object detection and categorization algorithm.
In this repository, you'll find folders for our webscraper, basic object detection, annotation and exploratory notebooks.