| Abstract |
As the development of the smart society advances, computer vision has become a crucial technology supporting its progress. However, outdoor visual algorithms often face challenges in adverse weather conditions and complex lighting, such as rain, fog, and shadows, leading to a decline in the performance of intelligent systems reliant on these algorithms. This presentation will discuss how to utilize lighting information to achieve matching between target objects and shadow instances, showcasing practical applications in the fields of image and video editing. Additionally, the speaker will explore the use of weather models to render image training data under adverse weather conditions and design image restoration algorithms based on physical priors to enhance the clarity of images and videos in challenging weather. Finally, the presentation will highlight research progress in developing a universal visual model for handling various adverse weather conditions. |
| About the speaker |
Dr. Xiaowei Hu is presently a research scientist at the Shanghai Artificial Intelligence Laboratory, specializing in deep learning and low-level computer vision research. He earned his Ph.D. in Computer Science and Engineering from the Chinese University of Hong Kong and has authored over 40 academic papers in top journals and conferences in computer vision, including TPAMI, CVPR, ICCV. Dr. Hu has been acknowledged in the National High-level Young Talents Program of China and listed among the World’s Top 2% Scientists by Stanford University.
|
