Developing an Object Detection Model for Pedestrian Detection: Applications in Traffic Safety and Human-Robot Interactions

EasyChair Preprint 12518

Abstract

Object detection is a fundamental task in computer vision, and developing models to detect pedestrians has significant practical implications in various domains, including traffic safetyandhuman-robot interactions. This abstract presents an overview of a straightforward computer vision project aimed at developing an object detection model specifically designed for pedestriandetection. The objective of this project is to design and train an accurate and efficient object detectionmodel capable of identifying pedestrians in images or video streams. The model is developedusing deep learning techniques, leveraging state-of-the-art architectures such as convolutional neural networks (CNNs) and region-based convolutional neural networks (R-CNNs). The model is trained on large-scale datasets containing annotated pedestrian images, enabling it to learndiscriminative features and spatial relationships necessary for pedestrian detection. The proposed model's applications are twofold. Firstly, in the context of traffic safety, the object detection model can be deployed in intelligent transportation systems and autonomous vehiclesto enhance pedestrian detection capabilities. By accurately identifying pedestrians in real-time, traffic accidents can be mitigated, and pedestrian safety can be improved.

Keyphrases: Robotics, Technology, computer

@booklet{EasyChair:12518,
  author    = {Favour Olaoye and Kaledio Potter},
  title     = {Developing an Object Detection Model for Pedestrian Detection: Applications in Traffic Safety and Human-Robot Interactions},
  howpublished = {EasyChair Preprint 12518},
  year      = {EasyChair, 2024}}