Redirigiendo al acceso original de articulo en 22 segundos...
Inicio  /  Information  /  Vol: 13 Par: 10 (2022)  /  Artículo
ARTÍCULO
TITULO

Improving Performance in Person Reidentification Using Adaptive Multiple Loss Baseline

Zhongmiao Huang    
Liejun Wang    
Yongming Li    
Anyu Du and Shaochen Jiang    

Resumen

Currently, deep learning is the mainstream method to solve the problem of person reidentification. With the rapid development of neural networks in recent years, a number of neural network frameworks have emerged for it, so it is becoming more important to explore a simple and efficient baseline algorithm. In fact, the performance of the same module varies greatly in different positions of the network architecture. After exploring how modules can play a maximum role in the network and studying and summarizing existing algorithms, we designed an adaptive multiple loss baseline (AML) with a simple structure but powerful functions. In this network, we use an adaptive mining sample loss (AMS) and other modules, which can mine more information from input samples at the same time. Based on triplet loss, AMS loss can optimize the distance between the input sample and its positive and negative samples and protect structural information within the sample. During the experiment, we conducted several group tests and confirmed the high performance of AML baseline via the results. AML baseline has outstanding performance in three commonly used datasets. The two indicators of AML baseline on CUHK-03 are 25.7% and 26.8% higher than BagTricks.

 Artículos similares

       
 
Junyao Zhang, Hao Zhan and Baigang Mi    
The submerged inlet exhibits good stealth characteristics and lower drag, but it has a low total pressure recovery coefficient and high distortion rate, which limits its widespread application. This paper proposes a vortex diverter aimed at enhancing the... ver más
Revista: Aerospace

 
Razvan Daniel Zota, Ionu? Alexandru Cîmpeanu, Denis Alexandru Dragomir and Mihai Adrian Lungu    
Sustainable development, smart waste management, and circular economy principles are paramount to the significant worldwide trend of smart city-related research and projects. The basic hypothesis of our research is that artificial intelligence (AI)-based... ver más
Revista: Applied Sciences

 
Chunling Wang, Tianyi Hang, Changke Zhu and Qi Zhang    
The Czech Republic is one of the countries along the Belt and Road Initiative, and classifying land cover in the Czech Republic helps to understand the distribution of its forest resources, laying the foundation for forestry cooperation between China and... ver más
Revista: Applied Sciences

 
Jingyuan Liang, Shuiqing Lin and Xizheng Ke    
Four-level pulse amplitude modulation (PAM4) can transmit more information in the same symbol interval, effectively improving the information transmission rate and frequency band utilization of visible light communication (VLC). This paper proposes a met... ver más
Revista: Applied Sciences

 
Yue Zha, Yuanzhi Ke, Xiao Hu and Caiquan Xiong    
Named entity recognition (NER) is particularly challenging for medical texts due to the high domain specificity, abundance of technical terms, and sparsity of data in this field. In this work, we propose a novel attention layer, called the ?ontology atte... ver más
Revista: Applied Sciences