Main Article Content
Vehicle re-identification is a challenging task due to intra-class variability and inter-class similarity across non-overlapping cameras. To tackle these problems, recently proposed methods require additional annotation to extract more features for false positive image exclusion. In this paper, we propose a model powered by adaptive attention modules that requires fewer label annotations but still out-performs the previous models. We also include a re-ranking method that takes account of the importance of metadata feature embeddings in our paper. The proposed method is evaluated on CVPR AI City Challenge 2020 dataset and achieves mAP of 37.25% in Track 2.
Comments from Mentors
Quang Truong has undoubtedly developed his passion for learning through spreading knowledge, introducing the out-of-class experience to his classmates, consequently fostering a supportive and dynamic classroom environment. His knowledge of programming, data science, and machine learning is a significant advantage to my research team. His weekly proposals for improvements and explorations contribute substantially to the research progress. He is a resourceful assistant to identify and improve problems of our models. Quang also excels in collaborating with everyone and displays a high level of commitment.
With his passion and hard work, Quang has led our research team under my supervision to have a paper published by Elsevier in Procedia Computer Science. This work about vehicle re-identification through machine learning shows his latest achievement.