Multi-Pedestrian Detection using Hybrid ML Algorithms for Autonomous Vehicles

Forecasting vulnerable road user behavior is a prerequisite for the real-world implementation of Autonomous Driving Systems (ADS). The purpose of a pedestrian crossing should be detected instantaneously, particularly while driving in towns. This paper aims to detect multiple pedestrians and other automobiles while driving specifically on Indian Roads in real-time. Recent research suggests that vision-based models utilizing deep neural networks are useful for this purpose. For this paper we aim to develop an end-to-end pedestrian intention detection architecture that works well both during the day and at night. The main approach for the project is based on bounding boxes for object identification. using various deep learning techniques like YOLOv3, Darknet-53 and YOLOv7.