Search

Your search keyword '"YAO, Ziwei"' showing total 312 results
Start Over Author "YAO, Ziwei"
312 results on '"YAO, Ziwei"'

12. Cross-modal Scene Graph Matching for Relationship-aware Image-Text Retrieval

Author

Wang, Sijin, Wang, Ruiping, Yao, Ziwei, Shan, Shiguang, and Chen, Xilin

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Image-text retrieval of natural scenes has been a popular research topic. Since image and text are heterogeneous cross-modal data, one of the key challenges is how to learn comprehensive yet unified representations to express the multi-modal data. A natural scene image mainly involves two kinds of visual concepts, objects and their relationships, which are equally essential to image-text retrieval. Therefore, a good representation should account for both of them. In the light of recent success of scene graph in many CV and NLP tasks for describing complex natural scenes, we propose to represent image and text with two kinds of scene graphs: visual scene graph (VSG) and textual scene graph (TSG), each of which is exploited to jointly characterize objects and relationships in the corresponding modality. The image-text retrieval task is then naturally formulated as cross-modal scene graph matching. Specifically, we design two particular scene graph encoders in our model for VSG and TSG, which can refine the representation of each node on the graph by aggregating neighborhood information. As a result, both object-level and relationship-level cross-modal features can be obtained, which favorably enables us to evaluate the similarity of image and text in the two levels in a more plausible way. We achieve state-of-the-art results on Flickr30k and MSCOCO, which verifies the advantages of our graph matching based approach for image-text retrieval., Comment: Accepted by WACV 2020

Published
2019

21. Shipborne oceanic high-spectral-resolution lidar for accurate estimation of seawater depth-resolved optical properties

Author

Zhou, Yudi, Chen, Yang, Zhao, Hongkai, Jamet, Cédric, Dionisi, Davide, Chami, Malik, Di Girolamo, Paolo, Churnside, James H., Malinka, Aleksey, Zhao, Huade, Qiu, Dajun, Cui, Tingwei, Liu, Qun, Chen, Yatong, Phongphattarawat, Sornsiri, Wang, Nanchao, Chen, Sijie, Chen, Peng, Yao, Ziwei, Le, Chengfeng, Tao, Yuting, Xu, Peituo, Wang, Xiaobin, Wang, Binyu, Chen, Feitong, Ye, Chuang, Zhang, Kai, Liu, Chong, and Liu, Dong

Published
2022
Full Text
View/download PDF

31. Deep Separation Between In(III) and Fe(III) Ions by Regulating the Lewis Basicity of Adsorption Sites on Electrospun Fibers.

Author

Li, Min, Zhang, Lin, Xu, Damin, Cheng, Nianshou, Wang, Mingyue, Shao, Penghui, Yang, Liming, Yao, Ziwei, Zhao, Chun, Feng, Jian, and Luo, Xubiao

Subjects
LEWIS basicity, BRONSTED acids, IONS, LEWIS acidity, ADSORPTION (Chemistry), METAL ions
Abstract

The deep separation between In(III) and Fe(III) ions is considered as an important technological challenge in the engineering fields. The In(III) ions are closer to the soft acid in terms of Lewis acidity due to the smaller polarizability and greater deformability than the Fe(III) ions, which facilitates the possibility of separating them. Herein, the Lewis basicity of the N‐containing adsorption sites is regulated by the fluorine‐containing group to better match with the In(III) ions rather than Fe(III) ions, causing a significant discrepancy in the binding of the adsorption sites to them. Moreover, targeted programmed desorption techniques based on the potential energy differences of the adsorption sites can achieve the deep separation of In(III) and Fe(III) ions, and reach an exceptional concentration of 20.7 mmol L−1 for In(III) ions in the initial desorption solution with extremely high mass ratio (mIn/mFe = 151.7). This work provides a novel technique for in‐depth separation of In(III) and Fe(III) ions from mixed systems, and a new strategy for separating metal ions with similar or close properties. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

32. Potential of Mie–Fluorescence–Raman Lidar to Profile Chlorophyll a Concentration in Inland Waters

Author

Zhao, Hongkai, primary, Zhou, Yudi, additional, Wu, Hongda, additional, Kutser, Tiit, additional, Han, Yicai, additional, Ma, Ronghua, additional, Yao, Ziwei, additional, Zhao, Huade, additional, Xu, Peituo, additional, Jiang, Chengchong, additional, Gu, Qiuling, additional, Ma, Shizhe, additional, Wu, Lingyun, additional, Chen, Yang, additional, Sheng, Haiyan, additional, Wan, Xueping, additional, Chen, Wentai, additional, Chen, Xiaolong, additional, Bai, Jian, additional, Wu, Lan, additional, Liu, Qun, additional, Sun, Wenbo, additional, Yang, Suhui, additional, Hu, Miao, additional, Liu, Chong, additional, and Liu, Dong, additional

Published
2023
Full Text
View/download PDF

45. Potential of Mie–Fluorescence–Raman Lidar to Profile Chlorophyll aConcentration in Inland Waters

Author

Zhao, Hongkai, Zhou, Yudi, Wu, Hongda, Kutser, Tiit, Han, Yicai, Ma, Ronghua, Yao, Ziwei, Zhao, Huade, Xu, Peituo, Jiang, Chengchong, Gu, Qiuling, Ma, Shizhe, Wu, Lingyun, Chen, Yang, Sheng, Haiyan, Wan, Xueping, Chen, Wentai, Chen, Xiaolong, Bai, Jian, Wu, Lan, Liu, Qun, Sun, Wenbo, Yang, Suhui, Hu, Miao, Liu, Chong, and Liu, Dong

Abstract

Vertical distribution of phytoplankton is crucial for assessing the trophic status and primary production in inland waters. However, there is sparse information about phytoplankton vertical distribution due to the lack of sufficient measurements. Here, we report, to the best of our knowledge, the first Mie–fluorescence–Raman lidar (MFRL) measurements of continuous chlorophyll a(Chl-a) profiles as well as their parametrization in inland water. The lidar-measured Chl-aduring several experiments showed good agreement with the in situ data. A case study verified that MFRL had the potential to profile the Chl-aconcentration. The results revealed that the maintenance of subsurface chlorophyll maxima (SCM) was influenced by light and nutrient inputs. Furthermore, inspired by the observations from MFRL, an SCM model built upon surface Chl-aconcentration and euphotic layer depth was proposed with root mean square relative difference of 16.5% compared to MFRL observations, providing the possibility to map 3D Chl-adistribution in aquatic ecosystems by integrated active–passive remote sensing technology. Profiling and modeling Chl-aconcentration with MFRL are expected to be of paramount importance for monitoring inland water ecosystems and environments.

Published
2023
Full Text
View/download PDF

48. Use of QSAR and SSD methods on deriving predicted no-effect concentrations in seawater and sediment for ten individual parent- and alkyl-PAHs and a case study on the assessment of their ecological risks from the Dalian Bay, China

Author

Juying Wang, Wang Ying, Liang Liu, Jingli Mu, Dian Zhang, Jin Fei, Xing Liu, Yi Cong, and Yao Ziwei

Subjects
Pollutant, 010504 meteorology & atmospheric sciences, Ecology, Aquatic ecosystem, Sediment, Aquatic Science, 010502 geochemistry & geophysics, Oceanography, 01 natural sciences, Environmental science, Marine ecosystem, Seawater, Ecosystem, Risk assessment, Bay, 0105 earth and related environmental sciences
Abstract

Parent and alkylated polycyclic aromatic hydrocarbons (alkyl-PAHs), which are a class of important toxic components of crude oil especially in the marine environment, exhibit adverse effects on aquatic life and potentially pose a human health risk. However, the lack of chronic toxicity data is one of the hindrances for alkyl-PAHs when assessing their ecological risks. In this study, predicted no-effect concentrations (PNECs) in seawater and marine sediment for ten parent- and alkyl-PAHs were derived by applying species sensitivity distributions (SSDs) and quantitative structure-activity relationships (QSARs). The local area, Dalian Bay, where an oil-spilled accident happened in 2010, was chosen as a case site to assess ecological risks for ten PAHs in surface seawaters and marine sediments. Their PNECs in seawater and sediment for protecting aquatic organisms in marine ecosystems were calculated and recommended in the range of 0.012–2.79 μg/L and 48.2–1 337 ng/g (dry weight), respectively. Overall, the derived PNECs for the studied PAHs in seawater and marine sediment were comparable to those obtained by classical methods. Risk quotient results indicate low ecological risks to ecosystems for ten parent- and alkyl-PAHs in surface seawaters and surface sediments from the Dalian Bay. These findings provide a first insight into the PNECs and ecological risks of alkyl-PAHs, emphasizing the role of the computational toxicology in ecological risk assessments. The use of QSARs has been identified as a valuable tool for preliminarily assessing ecological risks of emerging pollutants, being more predictable of real exposure scenarios for risk assessment purposes.

Published
2020

Catalog

Books, media, physical & digital resources
See catalog results