This project was completed as a part of the Honors portion of the Convolutional Neural Networks Course on Coursera.
Credit to DeepLearning.AI and the Coursera platform for providing the course materials and guidance.
As a part of my self-driving car project, I aim to develop a car detection system, which serves as a critical component of the overall endeavor. The images, provided by drive.ai, have been labeled with bounding boxes encompassing each car found in the scenes. My objective is to implement YOLO ("You Only Look Once") for object detection, specifically applying it to car detection.
The dataset comprises images with bounding box annotations, and I will explore two representations for the class label "c," either as an integer from 1 to 80 or as an 80-dimensional vector with 1 as one component and 0 for the rest. Throughout this exercise, I will delve into YOLO's object detection mechanism and subsequently implement it for car detection.
Since the YOLO model demands considerable computational resources for training, I will leverage pre-trained weights that are already loaded for convenient use. This will enable me to efficiently apply the YOLO model and accomplish accurate car detection in real-world scenarios while making significant strides towards achieving my goal of building a reliable self-driving car.
