Vehicle Image
Vehicle image analysis is a rapidly evolving field focused on extracting meaningful information from images of vehicles for various applications, primarily in intelligent transportation systems (ITS). Current research emphasizes improving vehicle re-identification (ReID) accuracy across diverse conditions using deep learning models like ResNet, Vision Transformers, and Swin-Transformers, often incorporating optimized regions of interest (ROIs) for enhanced feature extraction. These advancements are crucial for improving traffic monitoring, safety, and management, with applications ranging from automated number plate recognition to driver intention prediction and fine-grained vehicle classification for improved traffic flow analysis.
Papers
Fine-Grained Vehicle Classification in Urban Traffic Scenes using Deep Learning
Syeda Aneeba Najeeb, Rana Hammad Raza, Adeel Yusuf, Zamra Sultan
Automated Approach for Computer Vision-based Vehicle Movement Classification at Traffic Intersections
Udita Jana, Jyoti Prakash Das Karmakar, Pranamesh Chakraborty, Tingting Huang, Dave Ness, Duane Ritcher, Anuj Sharma