A5100348259- Financial Risk and Volatility Modeling
- Complex Systems and Time Series Analysis
- Video Surveillance and Tracking Methods
- Stock Market Forecasting Methods
- Image and Signal Denoising Methods
- Market Dynamics and Volatility
- Human Pose and Action Recognition
- Anomaly Detection Techniques and Applications
- Financial Markets and Investment Strategies
- Advanced Image and Video Retrieval Techniques
- Monetary Policy and Economic Impact
- Energy Efficient Wireless Sensor Networks
- Transportation Planning and Optimization
- Organic Light-Emitting Diodes Research
- Luminescence and Fluorescent Materials
- Consumer Market Behavior and Pricing
- Domain Adaptation and Few-Shot Learning
- Face recognition and analysis
- Traffic Prediction and Management Techniques
- Advanced Vision and Imaging
- 3D IC and TSV technologies
- Metalloenzymes and iron-sulfur proteins
- CO2 Reduction Techniques and Catalysts
- Energy Harvesting in Wireless Networks
- Customer churn and segmentation
Montpellier Business School
2021-2025
South China Agricultural University
2024
Donghua University
2024
National Yang Ming Chiao Tung University
2015-2023
Wuhan University
2023
Université de Montpellier
2021-2023
National Taiwan University
2016-2022
Guangdong University of Foreign Studies
2022
National Taipei University of Technology
2022
Google (United States)
2021
Pre-trained representations are becoming crucial for many NLP and perception tasks. While representation learning in has transitioned to training on raw text without human annotations, visual vision-language still rely heavily curated datasets that expensive or require expert knowledge. For vision applications, mostly learned using with explicit class labels such as ImageNet OpenImages. vision-language, popular like Conceptual Captions, MSCOCO, CLIP all involve a non-trivial data collection...
Age-related macular degeneration (AMD) is the leading cause of visual impairment among elderly in world. Early detection AMD great importance, as vision loss caused by this disease irreversible and permanent. Color fundus photography most cost-effective imaging modality to screen for retinal disorders. Cutting edge deep learning based algorithms have been recently developed automatically detecting from images. However, there are still lack a comprehensive annotated dataset standard...
Abstract We propose a new concept exploiting thermally activated delayed fluorescence (TADF) molecules as photosensitizers, storage units and signal transducers to harness solar thermal energy. Molecular composites based on the TADF core phenoxazine–triphenyltriazine (PXZ-TRZ) anchored with norbornadiene (NBD) were synthesized, yielding compounds PZDN PZTN two four NBD units, respectively. Upon visible-light excitation, energy transfer triplet state of occurred, followed by → quadricyclane...
This paper highlights the relation between shape of iron oxide (Fe3O4) particles and their magnetic sensing ability. We synthesized Fe3O4 nanocubes nanospheres having tunable sizes via solvothermal thermal decomposition synthesis reactions, respectively, to obtain samples in which volumes body diagonals/diameters were equivalent. Vibrating sample magnetometry (VSM) data showed that saturation magnetization (Ms) coercivity 100-225 nm cubic nanoparticles (MNPs) were, 1.4-3.0 1.1-8.4 times...
A family of new branched phosphine derivatives {Ph2N–(C6H4)n−}3P → E (E = O 1–3, n 1–3; S 4–6, Se 7–9, AuC6F5 1–3), which are the donor–acceptor type molecules, exhibit efficient deep blue room temperature fluorescence (λem 403–483 nm in CH2Cl2 solution, λem 400–469 solid state). Fine tuning emission characteristics can be achieved varying length aromatic oligophenylene bridge −(C6H4)n–. The pyramidal geometry central R3P fragment on one hand disrupts π-conjugation between branches to...
In this contribution, we design, implement and evaluate the pedagogical benefits of a novel interactive note taking interface (iVRNote) in VR for purpose learning reflection lectures. future environments, students would have challenges notes when they wear head mounted display (HMD). To solve problem, installed digital tablet on desk provided several tools to facilitate experience. Specifically, track stylus' position orientation physical world then render virtual stylus VR. other words, see...
Existing non-image-based solutions for detecting Phalaenopsis orchid diseases rely on time domain analysis of environmental conditions. These are suitable loosely controlled open farm fields. Conversely, the infrequent events in Internet Things (IoT)-based smart greenhouses, where conditions well-controlled. In tightly settings, standard methods fall short subtle changes during rare disease events. We observed that influence series temperature and humidity fungus within well-controlled...
A series of gold(I) alkynyl-diphosphine complexes (XC6H4C2Au)PPh2—spacer—PPh2(AuC2C6H4X); spacer = —C2(C6H4)nC2— (A1, n 2, X CF3; A2, OMe; A3, 3, A4, OMe), —(C6H4)n— (B5, B6, 4, OMe) were prepared, and their photophysical properties investigated. The luminescence behavior the titled compounds is dominated by diphosphine spacer, which serves as an emitting ππ* chromophore. exhibit dual emission, comprising low high energy bands triplet (phosphorescence) singlet (fluorescence) origins,...
The purpose of extractive summarization is to automatically select a number indicative sentences, passages, or paragraphs from the original document according target ratio and then sequence them form concise summary. In paper, we proposed use probabilistic latent topical information for spoken documents. Various kinds modeling structures learning approaches were extensively investigated. addition, capabilities verified by comparison with conventional vector space model semantic indexing...