EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Identity-Guided Spatial Attention for Vehicle Re-Identification

Kai Lv iD, Sheng Han, Youfang Lin

Sensors · 28.05.2023 · 23 (11) · 5152

DOI: 10.3390/s23115152 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 28.05.2023. Verfasst von Kai Lv und 2 weiteren Autor:innen. Beteiligte Einrichtung: Beijing Key Laboratory of Traffic Data Analysisand Mining, School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044…. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 70 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In vehicle re-identification, identifying a specific vehicle from a large image dataset is challenging due to occlusion and complex backgrounds. Deep models struggle to identify vehicles accurately when critical details are occluded or the background is distracting. To mitigate the impact of these noisy factors, we propose Identity-guided Spatial Attention (ISA) to extract more beneficial details for vehicle re-identification. Our approach begins by visualizing the high activation regions of a strong baseline method and identifying noisy objects involved during training. ISA generates an attention map to mask most discriminative areas, without the need for manual annotation. Finally, the ISA map refines the embedding feature in an end-to-end manner to improve vehicle re-identification accuracy. Visualization experiments demonstrate ISA’s ability to capture nearly all vehicle details, while results on three vehicle re-identification datasets show that our method outperforms state-of-the-art approaches.

Schlagwörter

Stichwörter der Autor:innen

attention mechanismdeep learningmachine learningvehicle detailsvehicle re-identification

Autor:innen und Institutionen

  • Beijing Key Laboratory of Traffic Data Analysisand Mining, School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China.

ORCID-Profile: Kai Lv

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Lv, K., Han, S., & Lin, Y. (2023). Identity-Guided Spatial Attention for Vehicle Re-Identification. Sensors, 23(11), 5152. https://doi.org/10.3390/s23115152

Vancouver

Lv K, Han S, Lin Y. Identity-Guided Spatial Attention for Vehicle Re-Identification. Sensors. 2023;23(11):5152. doi:10.3390/s23115152

BibTeX
@article{Lv2023,
  author = {Lv, Kai and Han, Sheng and Lin, Youfang},
  title = {Identity-Guided Spatial Attention for Vehicle Re-Identification},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {11},
  pages = {5152},
  doi = {10.3390/s23115152}
}
Kostenlos starten →