Low-Level Image Processing with Priors
数学专题报告
报告题目(Title)：Low-Level Image Processing with Priors
报告人(Speaker)：刘俊 （东北师范大学副教授)
地点(Place)：腾讯会议号 685-852-083
时间(Time)：2022 年 6 月 7日（周二） 19:00--20:00
邀请人(Inviter)：刘君
报告摘要
In this talk, I will introduce our recent work on two classic low-level image processing subjects: image deblurring and image enhancement. For the image deblurring, we proposed a surface-aware based term to smooth the inside regions segmented by L0 gradient term. The surface-aware strategy is from an intrinsic geometrical consideration. The superiority of our method in kernel estimation over the related methods is demonstrated by a series of experiments. For the image enhancement, we propose a real-time light correction method to recover the degraded scenes in the cases of sandstorms, underwater, and haze. We consider an intensity projection strategy to estimate the transmission. This strategy is motivated by a straightforward rank-one transmission prior. The complexity of transmission estimation is O(N) where N is the size of the single image. Comprehensive experiments on different types of weather/imaging conditions illustrate that our method outperforms competitively several state-of-the-art imaging methods in terms of efficiency and robustness.
主讲人简介
刘俊, 东北师范大学数学与统计学学院副教授。2015年博士毕业于电子科技大学, 博士期间，于2014年-2015年在美国加州大学洛杉矶分校（UCLA）交流学习。分别于2017年和2018年短期访问香港浸会大学与香港中文大学，曾主持中国博士后科学基金面上项目和国家自然科学基金青年项目。研究方向为底层图像处理问题的数学建模与算法设计，相关研究工作发表在一些国际期刊与会议，如IEEE TPAMI，IEEE TCI, Journal of Scientific Computing, Neurocomputing，Applied Mathematical Modeling，Information Sciences，JOSA A, CVPR，ECCV，ICIP等。
