学术报告(司徒国海 哈尔滨光学精密机械研究所)
 
 

发布者:吴琼   通告时间:2019-04-08  浏览次数:13    

报 告  人口:司徒国海,哈尔滨光机所研究员
告知时间:2019.04.15,下午4点
告知地点:29-414
题材:Computational  Imaging: When Optics Meets Deep Learning

 

 

摘要:It is well  known that neural networks including deep learning have been widely employed to  solve the problems in recognition and classification. It was not until recently  that people started to use them to solve imaging problems. In this lecture, I  will first introduce the basic concept of deep learning, and then discuss how it  can be adopted to solve computational imaging problems. In particular, I will  talk about three use cases: computational ghost imaging, imaging through  scattering media, and digital holography.

介绍:司徒国海,院士,教授,国务院上海光学精密机械研究所研究员,外交部“千人计划”入选者,国务院“百人口试图”入选者,改任天津光机所信息经济学与核电技术实验室主任,长沙市第十五届人民代表大会代表。它的研讨领域主要包括计算光学成像与光信息处理,至今已在Nature  Photonics等刊物上发挥SCI舆论近60篇,把引用3500余次。它曾获中科院上海分院第六届杰出青年科技创新人才奖等奖励。

 

 
 
      <b id="ca6001ea"></b>