Abstract

In this work, models for re-identification of pedestrians, based on deep learning, are studied. The models considered here use only images of pedestrians and do not require additional spatiotemporal information that can vary in different instances. A modification of the investigated models is proposed, based on the normalization of feature maps. In the best case, an assessment of the accuracy of the model demonstrates a 10% improvement over unmodified models.

© 2020 Optical Society of America

Deep learning-based cell identification and disease diagnosis using spatio-temporal cellular dynamics in compact digital holographic microscopy

Timothy O’Connor, Arun Anand, Biree Andemariam, and Bahram Javidi
Biomed. Opt. Express 11(8) 4491-4508 (2020)

Res-U2Net: untrained deep learning for phase retrieval and image reconstruction

Carlos Osorio Quero, Daniel Leykam, and Irving Rondon Ojeda
J. Opt. Soc. Am. A 41(5) 766-773 (2024)

Blood species identification based on deep learning analysis of Raman spectra

Shan Huang, Peng Wang, Yubing Tian, Pengli Bai, DaQing Chen, Ce Wang, JianSheng Chen, ZhaoBang Liu, Jian Zheng, WenMing Yao, JianXin Li, and Jing Gao
Biomed. Opt. Express 10(12) 6129-6144 (2019)

Vehicle identification using deep learning for expressway monitoring based on ultra-weak FBG arrays

Fang Liu, Yu Lei, Yu Xie, Xiaorui Li, Qiuming Nan, and Lina Yue
Opt. Express 31(10) 16754-16769 (2023)

Deep learning assisted non-contact defect identification method using diffraction phase microscopy

Subrahmanya Keremane Narayan, Allaparthi Venkata Satya Vithin, and Rajshekhar Gannavarpu
Appl. Opt. 62(20) 5433-5442 (2023)