Detection of Cyclists' Crossing Intentions for Autonomous Vehicles

Abadi, Arief Dharmawan; Gu, Yanlei; Goncharenko, Igor; Kamijo, Shunsuke

doi:10.1109/icce53296.2022.9730559

Cited by 8 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following the conventional definition used in the research of camera-based orientation estimation [ 19 , 20 ], cyclist body orientation is labelled in eight classes and head orientation of body orientation is labelled in three classes in this research. The body orientation number is defined along clockwise, starting from an orientation facing the LiDAR sensor.…”

Section: Cyclist Orientation Estimation Based On 2d and 3d Methodsmentioning

confidence: 99%

“…Raza et al presented an appearance-based pedestrian head-pose and full-body orientation prediction by using grayscale image and employing a deep learning mechanism [ 17 ]. Abadi et al proposed to estimate the cyclist head and body orientation using joined head map information generated from Openpose [ 18 ], and then used the joined head and body orientation to predict the crossing intention of cyclist [ 19 , 20 ]. In order to identify the cyclist heading and predict their intentions, Garcia et al proposed a multi-class detection with eight classes according to orientations and presented a performance comparison for cyclist detection and orientation classification between the main deep-learning-based algorithms reported in the literature, such as SSD, Faster R-CNN and R-FCN [ 21 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cyclist Orientation Estimation Using LiDAR Data

Chang

Goncharenko

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

It is crucial for an autonomous vehicle to predict cyclist behavior before decision-making. When a cyclist is on real traffic roads, his or her body orientation indicates the current moving directions, and his or her head orientation indicates his or her intention for checking the road situation before making next movement. Therefore, estimating the orientation of cyclist’s body and head is an important factor of cyclist behavior prediction for autonomous driving. This research proposes to estimate cyclist orientation including both body and head orientation using deep neural network with the data from Light Detection and Ranging (LiDAR) sensor. In this research, two different methods are proposed for cyclist orientation estimation. The first method uses 2D images to represent the reflectivity, ambient and range information collected by LiDAR sensor. At the same time, the second method uses 3D point cloud data to represent the information collected from LiDAR sensor. The two proposed methods adopt a model ResNet50, which is a 50-layer convolutional neural network, for orientation classification. Hence, the performances of two methods are compared to achieve the most effective usage of LiDAR sensor data in cyclist orientation estimation. This research developed a cyclist dataset, which includes multiple cyclists with different body and head orientations. The experimental results showed that a model that uses 3D point cloud data has better performance for cyclist orientation estimation compared to the model that uses 2D images. Moreover, in the 3D point cloud data-based method, using reflectivity information has a more accurate estimation result than using ambient information.

show abstract

Section: Cyclist Orientation Estimation Based On 2d and 3d Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Cyclist Orientation Estimation Using LiDAR Data

Chang

Goncharenko

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…All the mentioned datasets were obtained from moving vehicles on either urban or countryside roads, so that is why they contain information about the cyclist’s context. The orientation detection of a cyclist through CNN-based classifiers is addressed in [31] , [32] , training the proposed architectures from data sets created by the authors, whose cyclist’s orientation takes value every 45 degrees. However, in none of the cited datasets, the images of the cyclists were correlated with the cyclist’s acceleration and orientation data.…”

Section: Hardware In Contextmentioning

confidence: 99%

CYCLOPS: A cyclists’ orientation data acquisition system using RGB camera and inertial measurement units (IMU)

Arias-Correa,

Robledo,

Londoño

et al. 2024

HardwareX

View full text Add to dashboard Cite

“…Other approaches aimed at pedestrian and cyclist safety on public roads are presented in [Vial et al, 2023], with the use of mobile sensors for traffic tracking applications, and in [Abadi et al, 2022], [Fang and López, 2020] and [Ahmed et al, 2019] who propose prediction of cyclist behavior by autonomous vehicles or driver assistance systems, using deep neural networks, which investigate movement tracking and pose estimation. In [Pool et al, 2019] the authors propose a neural network that identifies contextual information for cyclist path prediction by autonomous vehicles…”

Section: Cyclist Detectionmentioning

confidence: 99%

Improved generalization of cyclist detection on security cameras with the OpenImages Cyclists dataset

E S Nardi,

Padilha,

Tadashi Kamaura

et al. 2023

JIDM

View full text Add to dashboard Cite

Most large public datasets containing cyclists for training detectors based on Deep Learning have annotations for bicycles and people, but not for cyclists. Even when it is not the case, the quality and quantity of the images are limited. To overcome these limitations, we propose the new OpenImages Cyclists dataset, built through the pre-selection of images from the OpenImages set and a new algorithm for semiautomatic generation of cyclist annotation aided by people and bicycle detectors. A cyclist detector trained with this dataset achieved identification rates up to 78% and 89% in two different sets of images obtained from security cameras at USP, Campus São Paulo - Capital.

show abstract

Detection of Cyclists' Crossing Intentions for Autonomous Vehicles

Cited by 8 publications

References 16 publications

Cyclist Orientation Estimation Using LiDAR Data

Cyclist Orientation Estimation Using LiDAR Data

CYCLOPS: A cyclists’ orientation data acquisition system using RGB camera and inertial measurement units (IMU)

Improved generalization of cyclist detection on security cameras with the OpenImages Cyclists dataset

Contact Info

Product

Resources

About