Please use this identifier to cite or link to this item: https://hdl.handle.net/11147/13220
Title: Cut-in maneuver detection with self-supervised contrastive video representation learning
Authors: Nalçakan, Yağız
Baştanlar, Yalın
01. Izmir Institute of Technology
01. Izmir Institute of Technology
Keywords: Contrastive representation learning
Driver assistance systems
Vehicle maneuver classification
Issue Date: 2023
Publisher: Springer
Abstract: The detection of the maneuvers of the surrounding vehicles is important for autonomous vehicles to act accordingly to avoid possible accidents. This study proposes a framework based on contrastive representation learning to detect potentially dangerous cut-in maneuvers that can happen in front of the ego vehicle. First, the encoder network is trained in a self-supervised fashion with contrastive loss where two augmented videos of the same video clip stay close to each other in the embedding space, while augmentations from different videos stay far apart. Since no maneuver labeling is required in this step, a relatively large dataset can be used. After this self-supervised training, the encoder is fine-tuned with our cut-in/lane-pass labeled datasets. Instead of using original video frames, we simplified the scene by highlighting surrounding vehicles and ego-lane. We have investigated the use of several classification heads, augmentation types, and scene simplification alternatives. The most successful model outperforms the best fully supervised model by ∼ 2% with an accuracy of 92.52%
URI: https://doi.org/10.1007/s11760-023-02512-3
https://hdl.handle.net/11147/13220
Appears in Collections:Computer Engineering / Bilgisayar Mühendisliği
Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Files in This Item:
File Description SizeFormat 
s11760-023-02512-3.pdf
  Until 2025-07-01
Article1.78 MBAdobe PDFView/Open    Request a copy
Show full item record



CORE Recommender

SCOPUSTM   
Citations

2
checked on Mar 1, 2024

Page view(s)

128
checked on Feb 26, 2024

Google ScholarTM

Check




Altmetric


Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.