Region-based MinCut Geodesic Models for Image Segmentation
数学专题报告
报告题目(Title):Region-based MinCut Geodesic Models for Image Segmentation
报告人(Speaker):陈达 山东省人工智能研究院
地点(Place):后主楼1124
时间(Time):2025年7月3日 (周四) 下午13:30-14:30
邀请人(Inviter):段玉萍
报告摘要
Geodesic models have been used for long to address the problems of image seg- mentation, where the features of interest, i.e, object boundaries, can be mod- eled as geodesic paths. As an important advantage, the geodesic model is ca- pable of finding the global minimum of a well-designed curve energy, thus can avoid unexpected local minima. The solutions to the segmentation problems can be regarded as a way to find a simple closed curve globally minimizing the associated curve energy, done by solving the corresponding Hamiltonian- Jacobi-Bellman PDE using the fast and efficient Fast Marching method.
In contrast to finding the global minimum of a simplified curve energy that only consists of edge-based features, we have recently established the relationship with the classical continuous min-cut problem, allowing to extend the geodesic model to cover region- and edge-based terms in conjunction with convexity shape priors. Through designing adequate geodesic metrics, we now are able to compute geodesic paths according to various active contours terms, involv- ing curvature penalization, region-based statistical term and shape prior con- straints. We will present the mathematical background as well as concrete ap- plications to biomedical and natural images.
主讲人简介
陈达,博士,副研究员,硕士生导师,山东省优秀青年基金获得者,2017年3月获得法国巴黎文理研究大学应用数学博士学位,2016年10月至2019年3月分别在巴黎多芬纳大学和巴黎国立眼科医院从事博士后研究工作。2019年5月通过高层次人才引进计划全职加入山东省人工智能研究院,并作为课题组负责人组建了智能图像分析团队,着力于解决医学图像分析和视觉计算领域存在的重点和难点问题,同时兼顾几何基础模型的研究和发展。主要研究方向为测地线模型的基础研究及其在计算机视觉和医学图像分析应用等。截至目前为止,申请人以第一作者身份在人工智能和应用数学专业的顶级期刊和主流会议上发表论文二十多篇,包括PNAS、TPAMI、IJCV、IEEE TIP、JMIV、CVPR、ICCV、BMVC等。现主持国家自然科学基金青年项目一项,山东省优秀青年科学基金一项,并作为骨干人员参与国家科技部重点研发计划,国家自然科学基金面项目等。