Search

Your search keyword '"Wang, Yinzhen"' showing total 207 results
Start Over Author "Wang, Yinzhen" Publication Year Range Last 10 years
207 results on '"Wang, Yinzhen"'

3. Unsupervised Temporal Video Grounding with Deep Semantic Clustering

Author

Liu, Daizong, Qu, Xiaoye, Wang, Yinzhen, Di, Xing, Zou, Kai, Cheng, Yu, Xu, Zichuan, and Zhou, Pan

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning
Abstract

Temporal video grounding (TVG) aims to localize a target segment in a video according to a given sentence query. Though respectable works have made decent achievements in this task, they severely rely on abundant video-query paired data, which is expensive and time-consuming to collect in real-world scenarios. In this paper, we explore whether a video grounding model can be learned without any paired annotations. To the best of our knowledge, this paper is the first work trying to address TVG in an unsupervised setting. Considering there is no paired supervision, we propose a novel Deep Semantic Clustering Network (DSCNet) to leverage all semantic information from the whole query set to compose the possible activity in each video for grounding. Specifically, we first develop a language semantic mining module, which extracts implicit semantic features from the whole query set. Then, these language semantic features serve as the guidance to compose the activity in video via a video-based semantic aggregation module. Finally, we utilize a foreground attention branch to filter out the redundant background activities and refine the grounding results. To validate the effectiveness of our DSCNet, we conduct experiments on both ActivityNet Captions and Charades-STA datasets. The results demonstrate that DSCNet achieves competitive performance, and even outperforms most weakly-supervised approaches., Comment: Accepted by AAAI2022

Published
2022

30. Conjugated layered and tunnel structure enables long-lifespan and high-rate sodium-ion battery cathodes.

Author

Gu, Chenzhang, Zhao, Enyue, Li, Na, Gao, Kesheng, Wu, Kang, Ran, Peilin, Fu, Mengmeng, Wu, Qi, Zhao, Jinkui, and Wang, Yinzhen

Abstract

A cathode material Na0.42Mn0.96Ti0.04O2 with conjugated layered and tunnel structures was synthesized via a convenient solid-state method. The designed conjugated structure enables superior electrochemical properties such as 93.9% capacity retention after 200 cycles at 1 C and excellent rate capability with a reversible capacity of 83.9 mAh g−1 at 20 C. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

40. Multicolor Mechanoluminescence in Sr2Ga2GeO7: Pr3+ for Stress Sensing and Anticounterfeiting.

Author

Xiao, Binli, Xiong, Puxian, Wu, Sheng, Jiang, Dongliang, Chen, Kang, Xiao, Yao, Shao, Peishan, and Wang, Yinzhen

Subjects
MECHANICAL energy, SLIDING friction, ENERGY conversion, THERMOLUMINESCENCE, LUMINESCENCE
Abstract

Mechanical luminescence (ML) is the conversion of mechanical energy into light energy when a material is subjected to a force. It is challenging to develop ML materials with tunable luminescent properties by single ion doping, which have broad application prospects in the fields of stress sensing and dynamic signature anticounterfeiting. In this study, a Pr3+ doped Sr2Ga2GeO7 tunable ML material is reported. Due to the traps resulting from unequal substitution of Pr3+–Sr2+ in [SrO6] heptahedra, the deep/shallow traps redistribution occurs towards a multicolor ML from red to green. In addition, even after continuous friction sliding force (15 N), the ML intensities can be recovered after ultraviolet pre‐irradiation, indicating that Sr2Ga2GeO7: Pr3+ has excellent ML emission stability. Based on thermoluminescence analysis, potential PersL and ML mechanisms are further elaborated by a lattice engineering (trap regulation) strategy. In all, the development of single‐ion doped multicolor ML materials may have crucial research value and show a variety of potential applications in the fields of anticounterfeiting and information encryption. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

43. Ratiometric mechanoluminescence in single Pr3+‐activated LiGa5O8.

Author

Wu, Sheng, Xiong, Puxian, Qin, Kexin, Xiao, Yao, Xiao, Binli, Chen, Kang, Jiang, Dongliang, and Wang, Yinzhen

Abstract

Mechanoluminescence (ML) materials have found potential applications in information storage, anti‐counterfeiting, and stress sensing. Conventional stress sensing based on absolute ML intensity is prone to significant mistakes owing to the unpredictability of measurement surroundings. However, implementing a ratiometric ML sensing technique may considerably ameliorate this issue. In this study, a single activator‐doped gallate material (LiGa5O8:Pr3+) is proposed to determine the relationship between the ML intensity and the change in local positional symmetry that occurs when the material is subjected to stress. The sensing reliability of the ML intensity ratio under different factors (Force; Content; Thickness and Materials) is systematically analyzed, where the factor that has the greatest effect on the proportional ML is the concentration, with the ML intensity asymmetry ratio decreasing from 1.868 to 1.300 varying concentration at constant stress. The colour‐resolved visualization of stress sensing is further realized, which opens a new path for a ratiometric ML‐based strategy to improve the reliability of stress sensing. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results