CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume 12506 > Article
Open Access Paper
28 December 2022 Reference set based metric learning method for person re-identification against overfitting problem
Chenhao Si
Author Affiliations +
Proceedings Volume 12506, Third International Conference on Computer Science and Communication Technology (ICCSCT 2022); 125066G (2022) https://doi.org/10.1117/12.2662666
Event: International Conference on Computer Science and Communication Technology (ICCSCT 2022), 2022, Beijing, China
Abstract
Similarity distance measurement is an important method in data classification, data recognition and other tasks, and has a very wide range of applications in machine learning, computer vision and other fields. However, there are model overfitting problems in complex data classification identification tasks in existing metric learning model. And those problems will negatively infect the accuracy and stability of the metric models. We study on person re-identification (person re-ID) task to design a robust similarity distance metric learning model based on a novel approach of overcoming over-fitting problem. The proposed method sets up a reference set based on training sample. Using the reference set and test images to form similar sample pairs, we can optimize the distribution information and projection feature. Finally, by testing on benchmark dataset, VIPeR, the experimental results validate the effectiveness of the proposed method. It achieves the best identification rates.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
Chenhao Si "Reference set based metric learning method for person re-identification against overfitting problem", Proc. SPIE 12506, Third International Conference on Computer Science and Communication Technology (ICCSCT 2022), 125066G (28 December 2022); https://doi.org/10.1117/12.2662666
PROCEEDINGS
 6 PAGES
DOWNLOAD PAPER SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Machine learning
Computer vision technology
Data modeling
Pattern recognition
Overfitting
Education and training
Statistical modeling
RELATED CONTENT
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top