Arnold, Eduardo, Dianati, Mehrdad, de Temple, Robert and Fallah, Saber (2021) Data for Cooperative perception for 3D object detection in driving scenarios using infrastructure sensors. [Dataset]

	Plain Text (Readme file) README.txt - Published Version Available under License Creative Commons Attribution 4.0. Download (1203b)
	Other (Pre-trained Pytorch 3DOD models for Tjunction and Roundabout) saved_models.tar.xz - Published Version Available under License Creative Commons Attribution 4.0. Download (132Mb)
	Other (Compressed dataset used for training and evaluation of Tjunction and roundabout scenarios) dataset.tar.xz - Published Version Available under License Creative Commons Attribution 4.0. Download (3336Mb)

Request Changes to record.

Abstract

3D object detection is a common function within the perception system of an autonomous vehicle and outputs a list of 3D bounding boxes around objects of interest. Various 3D object detection methods have relied on fusion of different sensor modalities to overcome limitations of individual sensors. However, occlusion, limited field-of-view and low-point density of the sensor data cannot be reliably and cost-effectively addressed by multi-modal sensing from a single point of view. Alternatively, cooperative perception incorporates information from spatially diverse sensors distributed around the environment as a way to mitigate these limitations. This article proposes two schemes for cooperative 3D object detection using single modality sensors. The early fusion scheme combines point clouds from multiple spatially diverse sensing points of view before detection. In contrast, the late fusion scheme fuses the independently detected bounding boxes from multiple spatially diverse sensors. We evaluate the performance of both schemes, and their hybrid combination, using a synthetic cooperative dataset created in two complex driving scenarios, a T-junction and a roundabout. The evaluation shows that the early fusion approach outperforms late fusion by a significant margin at the cost of higher communication bandwidth. The results demonstrate that cooperative perception can recall more than 95% of the objects as opposed to 30% for single-point sensing in the most challenging scenario. To provide practical insights into the deployment of such system, we report how the number of sensors and their configuration impact the detection performance of the system.

Item Type:	Dataset
Subjects:	Q Science > Q Science (General) T Technology > TA Engineering (General). Civil engineering (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering T Technology > TL Motor vehicles. Aeronautics. Astronautics
Divisions:	Faculty of Science, Engineering and Medicine > Engineering > WMG (Formerly the Warwick Manufacturing Group)
Type of Data:	Compressed dataset files and pre-trained pytorch models
Library of Congress Subject Headings (LCSH):	Automated vehicles, Computer vision, Image processing, Template matching (Digital image processing), Pattern recognition systems, Machine learning
Publisher:	University of Warwick, Warwick Manufacturing Group
Official Date:	13 October 2021
Dates:	Date Event 13 October 2021 Published 13 October 2021 Available 13 October 2021 Created
Status:	Not Peer Reviewed
Publication Status:	Published
Media of Output (format):	.tar.xz
Access rights to Published version:	Open Access (Creative Commons)
Copyright Holders:	University of Warwick
Description:	Cooperative 3D Object Detection using Infrastructure Sensors Dataset ==================================================================== Files Description ------------------ File: dataset.tar.xz Description: contains the Tar compressed dataset files used to train and evaluate the model on the T-junction and Roundabout scenarios. File: saved_models.tar.xz Description: contains the Tar compressed pre-trained pytorch weights for the T-junction and roundabout models. These were the models used for evaluation in the paper. Code Information ----------------- For more information on how to use these files, please read the official code repository at https://github.com/eduardohenriquearnold/coop-3dod-infra Disclaimer ----------- If you use this dataset, the pre-trained models or our code, please cite our work using @article{arnold_coop3dod, author={Arnold, Eduardo and Dianati, Mehrdad and de Temple, Robert and Fallah, Saber}, journal={IEEE Transactions on Intelligent Transportation Systems}, title={Cooperative Perception for 3D Object Detection in Driving Scenarios Using Infrastructure Sensors}, year={2020}, pages={1-13}, doi={10.1109/TITS.2020.3028424} }
Date of first compliant deposit:	13 October 2021
Date of first compliant Open Access:	13 October 2021
RIOXX Funder/Project Grant:	Project/Grant ID RIOXX Funder Name Funder ID UNSPECIFIED Jaguar Land Rover http://viaf.org/viaf/305209406 EP/N01300X/1 [EPSRC] Engineering and Physical Sciences Research Council http://dx.doi.org/10.13039/501100000266
Related URLs:	Other Other Related item in WRAP
Contributors:	Contribution Name Contributor ID Depositor Arnold, Eduardo 87202

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads