A Study of Autonomous Driving Object Detection Models Using Computer Vision and Deep Learning

Abstract

Object detection (OD) is a computer technique that enables localisation in computer vision, allowing the location and classification of objects in images/videos. With the help of deep learning (DL), OD has seen significant advancements, delivering higher accuracy with greater efficiency outcomes than traditional methods. This study provides an overview of work on object detection models. This study focuses especially on work in the field of autonomous driving over the last few years. The various models utilised in autonomous driving applications include LM-CNN-SVM, YOLOv4, and Bayesian Neural Networks (BNNs). In the discussed models, VGG16 yielded better results than the other models in this study. This study compares different models and algorithms in tabular form.