Yilin Zhao

ORCID: 0000-0001-5324-332X
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Visual Attention and Saliency Detection
  • Image Enhancement Techniques
  • Video Surveillance and Tracking Methods
  • Advanced Image and Video Retrieval Techniques
  • Spectroscopy and Laser Applications
  • Photovoltaic System Optimization Techniques
  • Advanced Neural Network Applications
  • Solar Radiation and Photovoltaics
  • Infrared Target Detection Methodologies
  • Robotics and Sensor-Based Localization
  • Water Quality Monitoring and Analysis
  • Remote Sensing and LiDAR Applications
  • Energy Load and Power Forecasting
  • Marine and coastal ecosystems
  • Advanced Optical Sensing Technologies
  • Persona Design and Applications

Shanghai Institute of Technology
2024-2025

Nanjing University of Posts and Telecommunications
2025

Beihang University
2024

State Grid Corporation of China (China)
2024

10.1109/icassp49660.2025.10887964 article EN ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2025-03-12

10.1016/j.jvcir.2024.104208 article EN Journal of Visual Communication and Image Representation 2024-06-01

10.1109/tim.2024.3450106 article EN IEEE Transactions on Instrumentation and Measurement 2024-01-01

10.1109/icme57554.2024.10687766 article EN 2022 IEEE International Conference on Multimedia and Expo (ICME) 2024-07-15

In order to solve the problem of low prediction accuracy caused by intermittency and uncertainty photovoltaic output, this paper, a parameter optimization variation mode decomposition (variation decomposition) based on sparrow search algorithm (SSA) is proposed. Combined model VMD long short-term memory neural network (LSTM). Firstly, Pearson correlation coefficient (PCC) was used analyze factors affecting PV output. Secondly, core parameters (k value penalty factor α) are automatically...

10.1117/12.3030652 article EN 2024-06-05

In this technical report, we target generating anthropomorphized personas for LLM-based characters in an online manner, including visual appearance, personality and tones, with only text descriptions. To achieve this, first leverage the in-context learning capability of LLMs generation by carefully designing a set system prompts. We then propose two novel concepts: mixture voices (MoV) diffusers (MoD) diverse voice appearance generation. For MoV, utilize text-to-speech (TTS) algorithms...

10.48550/arxiv.2311.06772 preprint EN other-oa arXiv (Cornell University) 2023-01-01
Coming Soon ...