Paper published in a book (Scientific congresses, symposiums and conference proceedings)
Poster: Lightweight Features Sharing for Real-Time Object Detection in Cooperative Driving
Hawlader, Faisal; Robinet, François; Frank, Raphaël
2023In 2023 IEEE Vehicular Networking Conference (VNC)
Peer reviewed
 

Files


Full Text
main.pdf
Author preprint (168.72 kB)
Request a copy

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Real-time Object Detection; Neural Network Quantization; Model Compression; V2X Communication
Abstract :
[en] In model partitioning for real-time object detection, part of the model is deployed on a vehicle, and the remaining layers are processed in the cloud. Model partitioning requires transmitting intermediate features to the cloud, which can be problematic, given that the latency requirements are strict. This paper addresses this issue by demonstrating a lightweight featuresharing strategy while investigating a trade-off between detection quality and latency. We report details on layer partitioning, such as which layers to split in order to achieve the desired accuracy.
Disciplines :
Computer science
Author, co-author :
Hawlader, Faisal  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > PI Engel
Robinet, François ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SEDAN
Frank, Raphaël ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SEDAN
External co-authors :
no
Language :
English
Title :
Poster: Lightweight Features Sharing for Real-Time Object Detection in Cooperative Driving
Publication date :
26 April 2023
Event name :
2023 IEEE Vehicular Networking Conference (VNC)
Event place :
Istanbul, Turkey
Event date :
26-04-2023 to 28-04-2023
Main work title :
2023 IEEE Vehicular Networking Conference (VNC)
Peer reviewed :
Peer reviewed
Available on ORBilu :
since 13 April 2023

Statistics


Number of views
134 (36 by Unilu)
Number of downloads
15 (15 by Unilu)

Scopus citations®
 
1
Scopus citations®
without self-citations
0

Bibliography


Similar publications



Contact ORBilu