Search

Your search keyword '"Shen, Fei"' showing total 3,911 results
Start Over Author "Shen, Fei"
3,911 results on '"Shen, Fei"'

67. GiT: Graph Interactive Transformer for Vehicle Re-identification

Author

Shen, Fei, Xie, Yi, Zhu, Jianqing, Zhu, Xiaobin, and Zeng, Huanqiang

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Transformers are more and more popular in computer vision, which treat an image as a sequence of patches and learn robust global features from the sequence. However, pure transformers are not entirely suitable for vehicle re-identification because vehicle re-identification requires both robust global features and discriminative local features. For that, a graph interactive transformer (GiT) is proposed in this paper. In the macro view, a list of GiT blocks are stacked to build a vehicle re-identification model, in where graphs are to extract discriminative local features within patches and transformers are to extract robust global features among patches. In the micro view, graphs and transformers are in an interactive status, bringing effective cooperation between local and global features. Specifically, one current graph is embedded after the former level's graph and transformer, while the current transform is embedded after the current graph and the former level's transformer. In addition to the interaction between graphs and transforms, the graph is a newly-designed local correction graph, which learns discriminative local features within a patch by exploring nodes' relationships. Extensive experiments on three large-scale vehicle re-identification datasets demonstrate that our GiT method is superior to state-of-the-art vehicle re-identification approaches., Comment: Accepted in IEEE TIP 2023

Published
2021

92. A Modeling Study on the Teaching System of Chinese Language in Colleges and Universities and the Cultivation Path of Bi-Creative Talents

Author

Shen Fei

Subjects
chinese language, dual-creative talent training, competency model, qfd model, reliability analysis, 97b20, Mathematics, QA1-939
Abstract

The trend of cultivating Chinese language dual-creative talents in colleges and universities is inevitable due to the increasing importance of national soft power. This paper studies the Chinese language teaching system and the cultivation path of dual-creative talents in colleges and universities using a modeling approach. The competency model’s iceberg model determines the dimensions and measurement items of dual-creative talents, verified through reliability analysis. At the same time, a QFD model for quality assurance of dual-creative talent cultivation in universities was established, and the importance of demand factors was calculated by analyzing the obtained results of each rating. Among the four factors, the degree of conformity was from the highest to the lowest: creative attitude, basic creative ability, creative expansion ability, professional knowledge, and skills, with rating values of 4.19, 4.12, 3.67, and 3.49, respectively. The empirical analysis of QFD showed that the highest weight of the demand level of development achievement of dual-creative talents cultivation reached 0.4066, followed by the basic conditions of teaching, which reached This study can provide some reference and help for the cultivation of Chinese language double-creative talents in colleges and universities, and help improve the quality of double-creative talents cultivation.

Published
2024
Full Text
View/download PDF

93. A Competitive Method to VIPriors Object Detection Challenge

Author

Shen, Fei, He, Xin, Wei, Mengwan, and Xie, Yi

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence
Abstract

In this report, we introduce the technical details of our submission to the VIPriors object detection challenge. Our solution is based on mmdetction of a strong baseline open-source detection toolbox. Firstly, we introduce an effective data augmentation method to address the lack of data problem, which contains bbox-jitter, grid-mask, and mix-up. Secondly, we present a robust region of interest (ROI) extraction method to learn more significant ROI features via embedding global context features. Thirdly, we propose a multi-model integration strategy to refinement the prediction box, which weighted boxes fusion (WBF). Experimental results demonstrate that our approach can significantly improve the average precision (AP) of object detection on the subset of the COCO2017 dataset.

Published
2021

94. Effect of High Temperature Storage Condition on Quality Deterioration of Peanut Oil and Protein

Author

LI Youhao, LIU Xiao, SHEN Fei, LIU Qiang, PEI Fei, MA Gaoxing, HU Qiuhui

Subjects
peanut storage, lipid oxidation, protein oxidation, oxidation-related enzymes, quality degradation, Food processing and manufacture, TP368-456
Abstract

Peanut is an important oilseed crop, susceptible to quality deterioration under harsh storage conditions because of its high oil content. In order to explore the effects of different storage temperatures on peanut quality under normal humidity conditions, in-shell peanuts were stored at 15, 25 or 35 ℃ for up to 30 weeks. Indicators of lipid oxidation: acid value (AV), peroxide value (PV), and malondialdehyde (MDA) content; indicators of protein oxidation: carbonyl, sulfhydryl, disulfide bond contents; and the activities of oxidation-related enzymes: lipase (LPS), lipoxygenase (LOX) and peroxidase (POD) were tested. The results showed that the AV increased from (0.48 ± 0.01) to (1.78 ± 0.02) mg/g and the carbonyl content from (3.19 ± 0.24) to (118.61 ± 6.41) μmol/g over the storage period at 25 ℃. Meanwhile, at the end of storage, the highest lipase and lipoxygenase activities were observed in the sample stored at 35 ℃, which were (59.00 ± 1.70) and (1 287.17 ± 98.45) U/g, respectively; the AV increased from (0.48 ± 0.01) to (3.15 ± 0.10) mg/g, the MDA content from 23.03 to 1 039.63 nmol/g, and the carbonyl content from (3.19 ± 0.24) to (124.86 ± 3.07) μmol/g. At 15 ℃, the highest lipase and lipoxygenase activities were only (41.60 ± 1.23) and (1 036.14 ± 34.49) U/g, respectively and the peroxide and carbonyl contents increased by 18.4 and 17.1 times after 30 days of storage, respectively, reaching much lower levels than those at the other temperatures. In conclusion, storage at normal humidity and at 15 ℃ can effectively suppress the oxidase activity at a low level at all time points, resulting in a low degree of oxidation of peanut oil and protein, while high ambient temperature increases the oxidase activity, and the higher the temperature, the higher the enzyme activity, and the higher the contents of primary and secondary oxidation products in oil, indicating a higher degree of peroxidation. Moreover, protein oxidation also occurs due to the oxidation of amino acid side chains and protein structural damage, ultimately leading to a decline in peanut quality.

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results