Detection of E-scooter Riders in Naturalistic Scenes

11/28/2021
by   Kumar Apurv, et al.
15

E-scooters have become ubiquitous vehicles in major cities around the world.The numbers of e-scooters keep escalating, increasing their interactions with other cars on the road. Normal behavior of an e-scooter rider varies enormously to other vulnerable road users. This situation creates new challenges for vehicle active safety systems and automated driving functionalities, which require the detection of e-scooter riders as the first step. To our best knowledge, there is no existing computer vision model to detect these e-scooter riders. This paper presents a novel vision-based system to differentiate between e-scooter riders and regular pedestrians and a benchmark data set for e-scooter riders in natural scenes. We propose an efficient pipeline built over two existing state-of-the-art convolutional neural networks (CNN), You Only Look Once (YOLOv3) and MobileNetV2. We fine-tune MobileNetV2 over our dataset and train the model to classify e-scooter riders and pedestrians. We obtain a recall of around 0.75 on our raw test sample to classify e-scooter riders with the whole pipeline. Moreover, the classification accuracy of trained MobileNetV2 on top of YOLOv3 is over 91 with precision and recall over 0.9.

READ FULL TEXT

page 2

page 3

page 4

page 5

research
10/09/2019

Intention Recognition of Pedestrians and Cyclists by 2D Pose Estimation

Anticipating the intentions of vulnerable road users (VRUs) such as pede...
research
08/17/2022

Blind-Spot Collision Detection System for Commercial Vehicles Using Multi Deep CNN Architecture

Buses and heavy vehicles have more blind spots compared to cars and othe...
research
03/06/2018

Where is my Device? - Detecting the Smart Device's Wearing Location in the Context of Active Safety for Vulnerable Road Users

This article describes an approach to detect the wearing location of sma...
research
05/09/2021

Interaction Detection Between Vehicles and Vulnerable Road Users: A Deep Generative Approach with Attention

Intersections where vehicles are permitted to turn and interact with vul...
research
05/17/2021

Learning to Automatically Catch Potholes in Worldwide Road Scene Images

Among several road hazards that are present in any paved way in the worl...
research
06/28/2017

A New Urban Objects Detection Framework Using Weakly Annotated Sets

Urban informatics explore data science methods to address different urba...
research
08/23/2018

Deep Learning Based Vehicle Make-Model Classification

This paper studies the problems of vehicle make & model classification. ...

Please sign up or login with your details

Forgot password? Click here to reset