Zobrazeno 1 - 10
of 1 954
pro vyhledávání: '"Yong Xu"'
Publikováno v:
IEEE Transactions on Multimedia. 25:267-279
Weakly supervised object detection (WSOD aims to train object detectors by using only image-level annotations. Many recent works on WSOD adopt multiple instance detection networks (MIDN, which usually generate a certain number of proposals and regard
Autor:
Zhijun Hu, Guanghai Liu, Raja Soosaimarian Peter Raj, Lilei Sun, Xianjing Cheng, Lian Wu, Jie Wen, Yong Xu
Publikováno v:
IEEE Transactions on Computational Social Systems. 9:1345-1358
Center loss is widely used as a supervision tool in deep learning method. However, the center loss also has some shortcomings, the most important of which is that it must be combined with softmax loss to run well. In this article, we sum up five shor
Publikováno v:
IEEE Transactions on Neural Networks and Learning Systems. 33:4979-4990
Heterogeneous palmprint recognition has attracted considerable research attention in recent years because it has the potential to greatly improve the recognition performance for personal authentication. In this article, we propose a simultaneous hete
Publikováno v:
IEEE Transactions on Industrial Informatics. 18:5171-5179
The continuous developments of urban and industrial environments have increased the demand for intelligent video surveillance. Deep learning has achieved remarkable performance for anomaly detection in surveillance videos. Previous approaches achieve
Publikováno v:
IEEE Transactions on Systems, Man, and Cybernetics: Systems. 52:3718-3730
Deep convolutional neural networks (CNNs) have been widely applied for low-level vision over the past five years. According to the nature of different applications, designing appropriate CNN architectures is developed. However, customized architectur
Autor:
Hui Wang, Yao Sun, Chun Huang, Yijia Cao, Yinglong Zhao, Yong Li, Li Jiang, Yong Xu, Ping Liu
Publikováno v:
IEEE Transactions on Transportation Electrification. 8:62-71
Although pulse charging strategy has many advantages, it is still controversial. The paradox is that it prolongs the charging time to eliminate the polarization voltage by the pulse interval, and the AC component in pulse current will increase the oh
Publikováno v:
Advances in Physiology Education. 46:21-26
The Internet and 5G era makes e-learning a vital part of modern education, and extensive evidence has shown that peer teaching and flipped classroom contribute to increased success in medical teaching. However, the applicability of these pedagogies i
Publikováno v:
IEEE Transactions on Multimedia. 24:728-739
Generating realistic pedestrian instances in a semi-supervised setting is promising but challenging due to the limited labeled data. We propose an unreliable-to-reliable instance translation model (Un2Reliab) conditioned on unreliable instances which
Publikováno v:
IEEE Transactions on Multimedia. 24:2975-2985
Learning effective generative models for natural image synthesis is a promising way to reduce the dependence of deep models on massive training data. This work focuses on Fine-Grained Image Synthesis (FGIS) in the semi-supervised setting where a smal
Autor:
Lin Zhou, Bingwen Zou, T. Mei, Xiaojuan Zhou, Xiaoman Tian, Yang Yu, You Lu, Min Quan, Libo Yan, Xuexi Yang, Yong Xu, Yongmei Liu, Jianxin Xue, Youling Gong, Qianyue Deng
Publikováno v:
International Journal of Radiation Oncology*Biology*Physics. 112:179-187
Purpose This study aimed to assess pathogen distributions and antimicrobial sensitivity characteristics in patients with non-small cell lung cancer (NSCLC) with severe radiation pneumonitis (SRP) and secondary infections. Methods and Materials Data f